oeoxvmaowucmc Acmwmnmr RIBONUCLEIC
ACID POLYMERASE or Psmnomoms PUTIDA: , f
STUDIES ON THE MECl-IANESM 0F Acmn. ‘ ~~

Thesis for the Degree of Ph. D.
MICHIGAN STATE umvm‘srw
GARY FLOYD GERARD
3972

rip—2...;

   
     

  

L! B R A R Y
Michigan State
University

This is to certify that the

thesis entitled

Deoxyribonucleic Acid-Dependent Ribonucleic Acid
Polymerase of Pseudomonas Putida: Studies on the
Mechanism of Action

 

presented by
Gary Floyd Gerard

has been accepted towards fulfillment
of the requirements for

PhoDo degree in BiOChemistry

 

 

BM
0

Major professor

Date “J C?! '1 l

0-7639

ABSTRACT

DEOXYRIBONUCLEIC ACID-DEPENDENT RIBONUCLEIC
ACID POLYMERASE OF PSEUDOMONAS PUTIDA:
STUDIES ON THE MECHANISM OF ACTION

BY

Gary Floyd Gerard

The objective of this research was to study the

mechanism of action of DNA—dependent RNA polymerase from

Pseudomonas putida. Three different aspects of the RNA

 

polymerase catalyzed synthesis of RNA were examined. First,
the requirement for a hydroxyl group at the 2'-position of
a ribonucleoside triphosphate substrate of RNA polymerase
was examined. Second, the requirement for a hydrogen or
hydroxyl group at the 2'-position of a polynucleotide tem-
plate of the enzyme was investigated. Third, the release
of the sigma subunit of RNA polymerase during DNA directed
RNA synthesis was studied using 35S-labeled RNA polymerase.
g. putida RNA polymerase can use 2'-Qfmethyladenosine
5'-triphosphate (AmTP) as a substrate for RNA synthesis. In
the DNA-directed synthesis of RNA with AmTP or an equimolar

mixture of AmTP and adenosine 5'-triphosphate (ATP) as the

 

Gary Floyd Gerard

adenyl substrates, a small amount of 2'-Qfmethyladenosine
5'—monophosphate (AmMP) was incorporated into RNA after a 60
minute incubation. This amounted to l to 2 pmoles of AmMP
for each pmole of RNA polymerase added to the reaction mix-
ture. AmMP-containing RNA synthesized in reaction mixtures
which contained AmTP as the only added adenyl substrate had
a sedimentation coefficient of less than 4 S as determined
by sucrose density gradient analysis. Abortive release of
the AmMP-containing RNA from the DNA-RNA polymerase-nascent
RNA ternary complex did not occur. AmMP-containing RNA
synthesized in reaction mixtures which contained both AmTP
and ATP as the adenyl substrates was heterogeneous in size
with most of the RNA having a sedimentation coefficient of
about 30 S. Degradation of the RNA product by alkaline
hydrolysis followed by alkaline phosphatase digestion
showed that 90% of the AmMP residues were located at the
3'-end of the RNA chain with the remaining 10% at the
5'-end or in the interior of the chain. The studies with
AmTP lead to the following conclusions. (1) A free 2'-
hydroxyl group is not required for binding of an adenyl
substrate by E. putig§_RNA polymerase or for subsequent
incorporation into RNA. (2) Following the incorporation

of AmMP into the 3'-end of the nascent RNA chain, the rate
of RNA chain growth is greatly reduced. The reduction in

the rate of RNA chain growth results in the accumulation

Gary Floyd Gerard

of nascent RNA chains with AmMP at the 3'—end. The rate-
limiting step in RNA synthesis becomes the addition of the
next nucleotide to the nascent RNA chain.

3. putida RNA polymerase can use 2'-gfmethyl-
polyuridylic acid (polyEUmJ) and 2'-Qfmethy1polycytidylic
acid as templates for the synthesis of polyadenylic acid
(poly[A]) and polyguanylic acid, respectively. No template
activity was detected with the purine-containing 2'-g—
methylated homopolymers, 2'—Q-methyladenylic acid (polyEAmJ)
and 2'—Q-methylinosinic acid. The poly(Am) strand of
either poly(Am) - poly(U) or poly(Am)- poly(Um) was not a
template for poly(U) synthesis, and did not prevent the
poly(U) or poly(Um) strand of the duplex from serving as
a template for poly(A) synthesis. The poly(Um) strand of
the duplex poly(Am)- poly(Um) was an effective template
for poly(A) synthesis, but the poly(Um) strand of poly(A)-
poly(Um) was not.
35S-labeled g. putida DNA-dependent RNA polymerase,
aZBB'o, was purified from cells that had been grown in a
minimal medium containing sodium [358]sulfate. The amount
of 358 in 8', B, and 0 relative to a was 3.6 to 3.6 to 2.2
to 1.0, respectively. A study of the release of the sigma
subunit of g. putida RNA polymerase was carried out follow-

ing the binding of enzyme to polynucleotides and during

DNA-directed RNA synthesis. Sucrose density gradient

Gary Floyd Gerard

centrifugation was the technique employed to assay for the

35S-labeled sigma. The subunits of 35S—labeled

release of
RNA polymerase present in protein peaks resolved on sucrose
gradients were identified by means of sodium dodecyl sulfate-
polyacrylamide gel electrophoresis. Binding of 35S—labeled
RNA polymerase to native DNA weakened the interaction

between sigma and core polymerase (dzBB') but did not

35S-labeled

result in the release of sigma. Binding of the
enzyme to polyadenylic acid, polycytidylic acid, and to
transfer RNA resulted in the release of sigma. Binding

of 35S—labeled RNA polymerase to an alternating copolymer

of deoxyadenosine and thymidine (poly[d(A—T)]), denatured
gh-l DNA, polythymidylic acid, and to polydeoxyadenylic

acid did not result in the release of sigma. Release of
sigma subsequent to the binding of enzyme to polydeoxy-
cytidylic acid and polyuridylic acid occurred in the absence
of manganese chloride but not in its presence. Sigma was
released from the enzyme-polynucleotide complex during
DNA-directed RNA synthesis. Within 3 minutes of incubation,
about 60% of the 35S-labeled RNA polymerase molecules
initiated RNA synthesis and formed a 200 mM KCl stable
complex with DNA and nascent RNA. All or almost all of
.these enzyme molecules released sigma. The other 40% of

the enzyme molecules did not form a 200 mM KCl stable

complex within 3 minutes. With longer times of incubation,

Gary Floyd Gerard

these enzyme molecules could slowly form a 200 mM KCl stable
complex but did not release sigma. The sedimentation
coefficient of g. putida sigma released during DNA-direction

RNA synthesis was 4.1 to 4.5 S.

DEOXYRIBONUCLEIC ACID-DEPENDENT RIBONUCLEIC
ACID POLYMERASE OF PSEUDOMONAS PUTIDA:

STUDIES ON THE MECHANISM OF ACTION

By

Gary Floyd Gerard

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1972

DED ICATION

To Kathy

ii

ACKNOWLEDGMENTS

My most sincere appreciation is expressed to
Dr. John A. Boezi for his guidance, selfless assistance,
and continual support given during all phases of my gradu-
ate training. Thanks are also given to Dr. Fritz Rottman
for his assistance in the design and evaluation of experi-
ments and in providing some of the biochemical reagents
used in my research.

Sincere appreciation is also expressed to the
other members of my Ph.D. guidance committee, Dr. William
C. Deal, Jr., Dr. Philip Filner, and Dr. Harold L. Sadoff.
I am indebted to Dr. J. C. Johnson for discussions and
for his collaborative efforts on the last section of
this thesis. Thanks are given to Howard C. Towle for
critical reading and evaluation of several sections of
this thesis. Appreciation goes to Dr. Kenneth Payne,
Mrs. Monique DeBacker, and Robert Blakesley for helpful
discussions.

I would like to thank the Department of Biochemistry
and especially the National Institutes of Health for finan—

cial support.

iii

TABLE OF CONTENTS

GENERAL INTRODUCTION . . . . . . . . . . . .

LITERATURE SURVEY . . . . . . . . . . . . .

Introduction . . . . . . . . . . . . .
Structure of RNA Polymerase . . . . . . . .
Binding . . . . . . . . . . . . . . .
Initial Bindin . . . . . . . . . . .
Enzyme-DNA Preinitiation Complex Formation . .
Rifampicin-Resistant Enzyme-DNA Complex
Formation . . . . . . . . . . .
The Role of RNA Polymerase Subunits in
Binding . . . . . . . . . . . .
Initiation . . . . . . . . . . . . . .
Nucleoside Triphosphate Binding Sites . . .
The Role of Sigma in Initiation . . . . .
The Role of Positive Control Factors in
Initiation . . . . . . . . . . .
Chain Elongation . . . . . . . . . . . .
Termination . . .
The Role of Rho in Termination . . . . . .
References . . . . . . . . . . . . . .

ARTICLE 1

2'-0-Methyladenosine 5'-Triphosphate. A Substrate
for Deoxyribonucleic Acid Dependent Ribonucleic
Acid Polymerase of Pseudomonas Putida, Gary F.
Gerard, Fritz Rottman, and John A. Boezi,
Biochemistry, lg, 1974 (l97l) . . . . . . .

 

ARTICLE 2

Template Activity of 2'-0-Methylpolyribonucleotides
with Pseudomonas Putida DNA-Dependent RNA Poly-
merase, Gary F. Gerard . . . . . . . . .

iv

Page

U1

OKDkalUT

l3
l4
14
15

18
20
21
22
24

3O

35

ARTICLE 3

Release of the Sigma Subunit of Pseudomonas
Putida Deoxyribonucleic Acid-Dependent
Ribonucleic Acid Polymerase, Gary F. Gerard

Page

52

GENE RAL INTRODUCT ION

DNA-dependent RNA polymerase is one of the most
complex enzymes, both structurally and functionally, yet
to be studied. The complete enzyme or holoenzyme from

Escherichia coli, Azotobacter vinelandii, and Pseudomonas

 

 

putida has five polypeptide subunits: one beta prime (8')
subunit, one beta (8) subunit, one sigma (0) subunit, and
two alpha (d) subunits. The subunits of P. putida holoenzyme
(a288'0) have the following molecular weights: 8', 165,000;
8, 155,000; 0, 98,000; and a, 44,000. In order for RNA
polymerase to catalyze the synthesis of a defined RNA
molecule from a DNA template, the enzyme must first initi—
ate RNA synthesis at a specific site on the DNA, it must
then faithfully c0py one of the strands of the DNA cistron
in accordance with the classic Watson-Crick rules of base
pairing, and the enzyme must finally terminate synthesis

at a second specific site on the DNA template. The sigma
subunit of RNA polymerase has the ability or imparts the
ability acting in concert with the other RNA polymerase
subunits, which collectively are called core enzyme (azBB'),
to recognize the specific DNA nucleotide sequences which

are signals for initiation of transcription. The core

enzyme alone is also capable of recognizing some terminating
DNA sequences, but other protein factors are required for
recognition by core enzyme of all termination signals.

The results presented in this thesis are concerned
with answering three different questions about the mechanism
of action of P. putida RNA polymerase. The first two
questions deal with RNA polymerase substrate and template
structural requirements. First, can RNA polymerase use a
nucleoside triphosphate as substrate which has a bulky
methoxy group rather than a hydroxyl group at the 2'-
position of the sugar moiety? Second, can RNA polymerase
use a polynucleotide as template which has a bulky methoxy
group rather than a hydrogen or hydroxyl group at the
2'—position of each sugar moiety? The third question
dealt with is outlined below.

A tenet regarding the RNA synthetic process which
is accepted as fact in current scientific thought and
communication, both in biochemical journals and textbooks,
is that the sigma subunit of RNA polymerase is released
during RNA synthesis in order to combine with other core
enzyme molecules which may subsequently bind to DNA to
initiate RNA synthesis. The initial and most frequently
cited experiment upon which the concept of sigma release
in yiE£2_is based, however, is both indirect and open to

other interpretations. No release of sigma in vivo has

as yet been demonstrated. We have attempted to answer the
question of whether or not sigma is released in_y££rg
during RNA synthesis with P. putida RNA polymerase by
using techniques which allow the physical separation of
released sigma from core enzyme and which permit identifi-
cation and quantitation of released sigma in a reaction
mixture.

This thesis is organized into four major divisions.
The first division is a literature survey in which much of
the current information on the mechanism of action of
bacterial DNA-dependent RNA polymerase has been summarized.
The second division is an article on 2'-gfmethyladenosine
5'-triphosphate as a substrate for P. putida RNA polymerase.
The article is presented in the form of a reprint from the
journal in which it was published. The third and fourth
divisions are articles presented in the format of a

scientific paper, with its own Abstract, Introduction,

 

 

Materials and Methods, Results, Discussion, and References

 

 

 

sections. Article Two deals with a study of 2'-gf
methylhomopolyribonucleotides as templates for P. putida
RNA polymerase. The contents of Article Two have been

submitted to Biochemical and Biophysical Research Com-

 

munications for publication. A preliminary report on

 

some of the data in this article was presented at the

69th Annual Meeting of the American Society for Microbiology

(Gary F. Gerard and J. A. Boezi, Bacteriological Proceedings,

 

69, 135 [1969]). Article Three is concerned with the

35

characterization of S-labeled RNA polymerase from P.

putida and the use of the enzyme to study the release of
sigma factor during DNA-directed RNA synthesis. The

l
contents of this article have been submitted to Biochemistry

 

for publication. A preliminary report on some of the data
in Article Three was presented at the 62nd Annual Meeting
of the American Society of Biological Chemists (Gary F.

Gerard, J. C. Johnson, and J. A. Boezi, Federation

 

Proceedings, 30, 1162 [1971]).

 

LITERATURE SURVEY

Introduction

 

In bacteria a single enzyme, DNA-dependent RNA
polymerase, is the enzyme directly reSponsible for the
synthesis of all types of cellular RNA. Genetic infor-
mation encoded in DNA is transcribed by RNA polymerase
into complementary RNA which may be translated into specific
proteins or may become the RNA of ribosomes or transfer RNA.
Since the first reports of the detection of DNA-dependent
RNA polymerase in bacteria in 1960, a prodigious literature
concerning the purification, the chemical, physical, and
structural characterization, and the elucidation of the
mechanism of action of bacterial RNA polymerase has ac-
cumulated. The extent and variety of topics treated in
the literature are beyond the scope of this survey. A
number of recent reviews and symposia concerning RNA
polymerase and transcription are available and should be
consulted for a comprehensive treatment of the literature
(1-8) .

RNA polymerase has been isolated and extensively
purified from a number of bacterial sources (1,2). The

enzymes from Escherichia coli (several strains) (9-11),

 

Azotobacter vinelandii (12-14), and Pseudomonas putida (15)

 

 

have been purified to homogeneity as determined by analysis
of protein using polyacrylamide gel electrophoresis. The
subunit composition of the enzyme from all three sources
has been shown to be essentially the same by means of
sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
RNA polymerase in the holoenzyme form has the following
polypeptide chain subunits: one beta prime subunit, 8';
one beta subunit, 8; two alpha subunits, a; and one sigma
subunit, 0. Holoenzyme from E. ggli can be separated into
two functional parts by phosphocellulose chromatography
(16): a core enzyme (dzBB') which is the basic machinery
for carrying out transcription, and a sigma (0) factor
which enables core to initiate RNA synthesis efficiently
at specific DNA sites. In addition to sigma factor, there
are several other protein factors (transcriptional factors)
which have a pronounced effect on transcription. These
transcriptional factors do not bind tightly enough to

core enzyme to purify with it.

This survey will be restricted to a discussion of
the enzyme from E, 331$ since it is the most highly studied
and consequently, the best known. In addition, emphasis
will be placed primarily on a discussion of the mechanism
of each step involved in the RNA synthetic process and

the role that the subunits of RNA polymerase and the

transcriptional factors play in specific steps. The steps
of RNA synthesis which will be delineated are: (1) binding
of RNA polymerase to the DNA template, (2) initiation of
RNA synthesis, which involves formation of the first
phosphodiester bond between the 5'—termina1 and the second
ribonucleotide, (3) propagation of synthesis or RNA chain
elongation, and (4) termination of synthesis and release

of enzyme and nascent RNA from the DNA template. Before
discussing the mechanism of RNA synthesis, a few more
details on the structure of E. coli RNA polymerase will

be outlined.

Structure of RNA Polymerase

 

E. ggli_RNA polymerase may be dissociated into its
individual polypeptide chain subunits by treatment with
denaturing agents such as sodium dodecyl sulfate (SDS)
or 8 M urea (17,18). The subunits can be separated
analytically by electrOphoresis on polyacrylamide gels
containing SDS or 8 M urea (17,18), or preparatively by
electrophoresis on cellulose acetate slabs containing 8 M
urea (19,21,24). The molecular weights of the subunits
have been determined by comparing mobilities of the SDS-
dissociated subunits on SDS-containing polyacrylamide gels
with the mobilities of polypeptides of known molecular
weight (20). The values obtained in several laboratories

are: 8', 150,000 to 165,000; 8, 145,000 to 155,000;

0, 85,000 to 95,000; and a, 39,000 to 41,000 (17,21,22).
The stoichiometry of the subunits in holoenzyme determined
by measuring relative band intensities on stained SDS-
polyacrylamide gels is d288'0 (22,23). Tryptic finger-
prints for each subunit have been obtained and they
indicate that B', B, o, and d have different amino acid
sequences (19,24). Core enzyme can be obtained from
holoenzyme by phosphocellulose chromatography (16) and

the stoichiometry of subunits in core is azBB' (17). The
minimal formula weights which can be calculated for core
enzyme and holoenzyme are 400,000 1 10% and 490,000 : 10%,
respectively. The molecule weight of the protomer form

of core enzyme as determined by sedimentation equilibrium
is 380,000 (22), which is in good agreement with the calcu-
lated minimal formula weight. As determined by sedimenta-
tion velocity and equilibrium experiments (22), both
holoenzyme and core enzyme exist as aggregates of their
protomer forms at low ionic strengths. Holoenzyme forms

a dimer of its protomer at ionic strengths below 0.12,

and core enzyme forms a complex containing at least six
core enzyme protomers at ionic strengths below 0.26.

Above ionic strengths of 0.12 and 0.26, respectively,

holoenzyme and core enzyme exist as protomers.

Binding

Initial Binding. The initial binding of RNA

 

polymerase to DNA is rapid and reversible (25-27), dependent
upon ionic strength (25,28,29), and inhibited by RNA (30)
and by the polyanion heparin (31). The number of enzyme
molecules which can be bound by a DNA template at low ionic
strength seems to be limited only by the amount of space
available on a particular template (32). Enzyme molecules
bound in a densely packed array on DNA, however, are bound
in a transcriptively nonproductive way (32,33). Experi-
ments involving a number of DNA templates including T4, T7,
A, and doubly-closed circular polyoma and papilloma DNA's
have demonstrated that under low ionic strength conditions
(ionic strength “ 0.1), enzyme molecules which can be
bound in a transcriptively productive manner are located

at comparatively widely spaced points along the DNA (25,
27,34,35). The quantities of enzyme bound correspond to
approximately one enzyme molecule in the dimer form to
every 2,000 to 2,500 A of DNA or every 750 DNA base pairs,
or 15 to 70 enzyme binding sites per DNA molecule depending
upon the size of the particular DNA species. This number
of binding sites is much larger than the actual number of
specific initiation sites or promoter sites thought to
exist on T4, T7, and A DNA (36-42). The preparations of

E. ggli_RNA polymerase used to study initial binding

probably rarely contained more than 50% of the stoichiometric

10

content of sigma (11,24). Indeed, sigma is not required
for the initial binding of RNA polymerase to DNA since both
holoenzyme and core enzyme bind to DNA in this manner with
equal facility (43).

Enzyme-DNA Preinitiation Complex Formation. Sub-

 

sequent to the initial binding of holoenzyme to T DNA, a

7
highly stable preinitiation complex can be formed between
holoenzyme and a small number of specific sites on the T7
DNA (44). Temperatures above 200 C are required to form

the complex, and at low ionic strength and 37° C the complex
has a half—life of 60 hours. The complex is much less
stable at higher ionic strength and lower temperatures.
Holoenzyme has also been observed to form similar pre-
initiation complexes with T4 DNA (24). Sigma is required
for the formation of these Specific preinitiation complexes,
since binding complexes formed between core enzyme and T7
DNA are much less stable and are not located at a small
number of specific sites (44). The sites on the T7 DNA

at which the preinitiation complexes are formed are thought

to be the same as the specific promoter sites on T at

7
which RNA synthesis is initiated in vitro and in vivo by
E. coli RNA polymerase (44). A possible reason which

might help to explain the descrepancy between the number

of initial binding sites and the smaller number of promoter

sites on bacteriOphage DNA has been suggested by Burgess (6).

11

RNA chains initiated by both core enzyme and holoenzyme
begin almost exclusively with purine nucleosides. The
second nucleoside is predominantly a pyrimidine with
holoenzyme, but with core enzyme the next nucleosides most
frequently are a run of one or more purines (45,46). The
polypurine runs with core enzyme suggest that it does not
start transcription entirely at random, but prefers to
bind to regions of the DNA template rich in pyrimidines.
This might also be true for holoenzyme since there is a
correlation between the bacteriophage DNA strand transcribed
preferentially by holoenzyme in yitrg and in_yiyg_and the
strand containing pyrimidine clusters (47,48). These
clusters are dispersed throughout the genome and their
spacing has been estimated as on the average one cluster 15
to 40 nucleotides long every 500 to 5000 DNA nucleotide pairs
(48-50). If only a fraction of the pyrimidine clusters
have specific initiation sequences contiguous to them
(1,48), it could help to eXplain the presence of a larger
number of initial binding sites than specific initiation
or promoter sites on DNA.

The strong temperature dependence of the formation
of the specific preinitiation complex between holoenzyme

and T7 DNA (44) or T4 DNA (24) suggests that a melting out

12

of the DNA helix occurs simultaneous with complex formation.
Sigma is thought to mediate this opening of the bound region
of the DNA helix since sigma is required for the formation
of the specific preinitiation complex. Indeed, core enzyme
appears to require a single-stranded or partially single-
stranded region in a DNA template at which to bind to allow
initiation of RNA synthesis. That is, polynucleotides such
as single-stranded DNA (16) and double-stranded DNA con-
taining or treated to produce single-strand breaks (16,51),
and Open structure double-helical DNA's such as poly[d(A-T)]
(11) and twisted closed—circular ¢X-174 RF (51) and fd RF
(45) are the only DNA's which serve as efficient templates
for core enzyme.

Rifampicin-Resistant Enzyme-DNA Complex Formation.

 

Rifampicin is an antibiotic which inhibits RNA synthesis
i2_yiyg_(52) and in yitrg_(53). The drug blocks a step

in RNA synthesis which occurs after binding of enzyme to
DNA but prior to chain initiation (5), and functions by
binding specifically to the 8 subunit of RNA polymerase
(54). Subsequent to formation of the stable preinitiation
complex described earlier, holoenzyme undergoes a change,
perhaps a conformational change, which results in the
formation of a rifampicin-resistant complex. This has
been shown to occur with T2, T4, and T7 DNA (24,36,55,56).

The two complexes can be distinguished because the stable

l3

preinitiation complex can be formed in the presence of
rifampicin (5) whereas formation of the rifampicin resistant
complex is blocked by rifampicin (36,56). The decay of the
rifampicin-resistant complex to rifampicin sensitivity in
the presence of the drug is two to three orders of magnitude
faster than the dissociation of the stable preinitiation
complex at the same ionic strength (36). Sigma is necessary
for the formation of the rifampicin-resistant complex with
intact double-stranded phage DNA (36,56). Enzyme bound to
rifampicin loses its ability to bind the initiating purine
nucleoside triphosphates even in the absence of DNA (57),
so that it has been suggested that the change in enzyme
associated with formation of the rifampicin-resistant com-
plex is necessary for correct binding of the initiating
purine nucleoside triphosphate and subsequent phosphodiester
bond formation with the second nucleoside triphosphate (6).
The Role of RNA Polymerase Subunits in Binding. The
subunit of core enzyme which by itself carries the specifi-
city required for binding to DNA is 8' (24). This conclusion
is based on the fact that B' can efficiently bind DNA to
membrane filters whereas 8, o, and a cannot (58,59). There
may be other DNA-binding sites, however, which require
interaction with other subunits in order to be eXposed.
Although sigma alone does not bind to DNA (60), the factor

is intimately involved in the selection of a specific

14

binding site by RNA polymerase and perhaps subsequent to
binding is required for local melting of the bound region

of the DNA helix.

Initiation

 

Nucleoside Triphosphate Binding Sites. RNA chain
initiation is usually defined as the oriented binding of
two nucleoside triphosphates to RNA polymerase followed
by phOSphodiester bond formation with the formation of a
dinucleoside tetraphosphate and the elimination of inor-
ganic perphosphate (5). It has been established for
some time that in 31:39 RNA synthesis is initiated almost
entirely with purine nucleotides (61-64), so that the
first nucleoside triphosphate which binds to enzyme is
a purine. The binding of nucleoside triphosphates to
enzyme has been studied through the use of kinetic analysis
(65), and by equilibrium dialysis and fluorescence quenching
(66,67). Based on these studies, a model has been proposed
which suggests the existence of at least two nucleoside
triphosphate-binding sites on RNA polymerase (57,67). One
site which has been designated as the initiation binding
site strongly prefers the binding of purine triphosphates.
This binding does not require divalent metal ion and is
inhibited by rifampicin. The second binding site, which
has been designated as the polymerization binding site,

exhibits approximately an equal affinity for all nucleoside

15

triphosphates, requires divalent metal ion, and is not
affected by rifampcin. The initiation site is a relatively
weak binding site while the polymerization site is a rela-
tively strong binding site. There is a tenfold difference
between the values for apparent substrate §m_at the two
sites. Some question has been raised, however, concerning
the tenfold difference in apparent 53:3, since for poly
[d(A—T)]-directed poly r(A-U) synthesis the apparent Eng
for ATP and UTP are the same measured either for initiation
or for polymerization (68). In this study, higher substrate
concentrations were required for initiation than chain
elongation because initiation was found to be a bimolecular
reaction and the ratelimiting step in poly r(A-U) synthesis,
while chain elongation was found to be first-order.

The Role of Sigma in Initiation. The sigma subunit

 

of E, goli_RNA polymerase has a dramatic influence on both
the efficiency and specificity with which the enzyme initi—
ates RNA synthesis. Initiation may be monitored by means

of a pyrophosphate exchange reaction catalyzed by RNA
polymerase (69) in which radioactive inorganic perphosphate
is incorporated into the ribonucleoside triphosphates
involved in initiation (13). Measurement of the pyro-
phosphate exchange reaction has shown that with T4 DNA

as the template, initiation of RNA synthesis is dependent

upon the presence of sigma (70). It was concluded from

16

this study that sigma is probably necessary for the forma—
tion of the first phosphodiester bond catalyzed by RNA
polymerase. In this regard, the apparent 5m values for
initiation of nucleoside triphosphates were found to be
higher in the absence of sigma then in its presence (45),
suggesting that sigma facilitates initiation. RNA chain
initiation can also be assayed by monitoring the DNA-
directed incorporation of y-32P-1abe1ed nucleoside
triphosphates into RNA, for only the first nucleotide
incorporated into the growing RNA chain retains its
triphosphate group (61,62). With intact double-stranded
bacteriophage DNA's such as T4, T7, and A as template, the
number of RNA chains initiated by RNA polymerase is some
5- to 75-fold greater in the presence of sigma than in
its absence (6).

Sigma is necessary for accurate initiation of RNA
synthesis at specific sites on the DNA template. Initiation
of RNA synthesis by E. goli_RNA polymerase in_vitro on T4
DNA (71,72), as well as T7 DNA (40,41), in the presence of
sigma is restricted to sites on one strand of the DNA which

are the sites utilized in vivo during the "early" period of

 

viral infection. In the absence of sigma, initiation of
RNA synthesis in vitro occurs at random sites on both
strands of the viral DNA. This has been demonstrated by

means of hydribization competition experiments with in vitro

17

and £2 yiyo RNA transcripts and separated strands of viral
DNA. Through the elegant work of Suguira, Okamoto, and
Takanami (45,73), a more direct demonstration has been
made of how sigma restricts initiation to specific sites
on DNA. These workers analyzed size, asymmetry, and 5'-
terminal sequences of RNA synthesized by E. ggii RNA
polymerase from fd phage replicative form DNA Ea vitro.
Holoenzyme initiates primarily three different RNA chains
of discrete size and initial sequence from one DNA strand.
In contrast, core enzyme transcribes both strands and pro—
duces RNA which has many initial sequences and is very
heterogeneous in size. These same analyses have been
performed on RNA transcripts synthesized from ¢80 DNA
(73,74), and again the results indicate that sigma restricts
initiation of RNA synthesis by E. go}; RNA polymerase to
specific sites on DNA.

Based mainly on the studies summarized in this
survey on the role of sigma in binding and initiation,
it has been hypothesized that sigma may play a dual role
in the RNA synthetic process: one in stabilizing enzyme-
DNA complexes at specific initiation (promoter) sites,
and the other in facilitating initiation of the RNA
chains at the specific sites (6). It is not known whether
actual recognition of the initiation sequence is carried

out by core enzyme or sigma. If the Specific information

18

for site selection is localized in sigma, then sigma would
direct the core polymerase to bind at the promoter site.

On the other hand core polymerase could identify a specific
nucleotide sequence and the role of sigma might be to act
as an allosteric effector to promote tight, site-specific
binding.

The Role of Positive Control Factors in Initiation.

 

The existence of a class of protein factors in bacteria
which function as positive control elements for initiation
of transcription of different general classes of RNA has
been hypothesized (76). These factors would act as second-
ary specificity determinants for promoter recognition by
RNA polymerase, the primary specificity determinant being
sigma. They would require the presence of sigma to func-
tion, and would be regulated by low molecular weight
effectors (75). It has been suggested that in addition
to the initiation site at which the sigma-directed inter-
action of holoenzyme with the DNA promoter takes place,
there is a secondary site located in the promoter region
which, if present, requires the binding of a positive-
control element for initiation of RNA synthesis (6). To
date, two such positive-control elements, wr and CAP, have
been isolated from E. ggli,

A positive-control factor or psi (w) factor has
been partially purified from E. goli_which is specific for

ribosomal RNA genes (75). This factor, wr, stimulates

V

‘m'-"
:J'
< u. .

I

 

 

19

transcription of E. goEE_DNA by holoenzyme but not by

core enzyme. In the absence of wr with E. golE_DNA as
template, less than 0.2% of the RNA synthesized by E, goli
holoenzyme $3 XEEEQ is ribosomal RNA (75,76). The RNA
synthesized in the presence of wr by holoenzyme is 30 to
40% ribosomal RNA (75). Guanosine tetraphosphate, ppGpp,
which is known to accumulate in stringent strains of

E. ggli_during amino acid starvation (77), and which is

lthought to directly inhibit ribosomal RNA synthesis i2 vivo

 

(78), has been tested for its effect on wr-stimulated RNA
synthesis. The nucleotide at a concentration of 0.4 mM
inhibits wr-stimulated RNA synthesis 50 to 65% while
inhibiting RNA synthesis by holoenzyme alone by only 10%
(79). No wr-directed ribosomal RNA synthesis was detected
at this ppGpp concentration.

A second positive control factor purified from
E,.gglE, catabolite gene-activating protein (CAP) (80)
or cyclic AMP receptor protein (CRP) (81), mediates the
effect of cyclic AMP which is required for the expression
of genes subject to catabolite repression (82). The action
of CAP in regulating the transcription of lactose operon-
<:ontaining DNA has been studied in a cell-free system
(83,84). In a purified system consisting of Egg operon-
<:ontaining DNA and core polymerase, the synthesis of $327

specific messenger RNA is stimulated only if CAP, cyclic

 

 

20

AMP, and sigma are all present. If lag repressor is added
to the complete system, about 80% of lggfspecific transcrip—
tion is blocked. This inhibition by repressor can be over-
come with synthetic inducers of the lag operon. Both cyclic
AMP and CAP are required for the formation of a rifampicin-

resistant preinitiation complex (36) which can make lac

___ I‘m
messenger RNA, indicating that both cyclic AMP and CAP f I
are required for binding of holoenzyme to Egg promoter. ;
Finally, it has been reported that CAP itself binds tightly ‘
to Egg-containing DNA (84). {FT
V

Chain Elongation

 

Growth of RNA chains occurs with 5' to 3' chemical
polarity with the sequential addition of nucleoside mono-
phosphates to the 3'-terminus of the nascent RNA chain
(61,62). The process involves classical Watson—Crick
base pairing between substrate ribonucleoside triphosphates
and the DNA template to produce a complementary RNA COpy
of one of the two strands in any region of the DNA tem—
plate. The overall rates of RNA chain elongation obtained
with current enzyme preparations and assay conditions
E2 vitro are comparable to those obtained E2 yiyo (42,85).

A change occurs in the enzyme at some point after
RNA chain initiation which makes the interaction between

enzyme and DNA resistant to high ionic strength (25,86).

21

With poly[d(A—T)] as the template, ATP or UTP alone does
not stabilize binding of holoenzyme to poly[d(A-T)] (87).
The dinucleotide primer, ApU, by itself does not stabilize
the complex against 0.1 M KCL and ATP must be added to form
ApUpA before stabilization occurs (88).

At some stage in E2 vitro chain elongation, sigma
is apparently released from the holoenzyme and core poly—
merase continues chain elongation (23). This has led to
the postulation of the sigma cycle in which sigma func-
tions in RNA synthesis only during initiation (23). What
triggers sigma release and the rationel for its release
are not clear. Several lines of evidence indicate that
sigma may be released i2 yiyg during RNA synthesis. RNA
polymerase-DNA complexes actively involved in RNA syn-
thesis have been isolated from E. ggii, and it has been
determined by SDS-polyacrylamide gel electrophoresis that
the complexes as obtained do not contain sigma (89). RNA
polymerase-A DNA complexes have been isolated from A-
infected E. EEli.WhiCh apparently do not contain detectable
amounts of sigma as determined by more indirect kinetic

analyses (90).

Termination

 

There are three aSpects of the process of termination
which are of interest: (1) the cessation of RNA synthesis

at a specific site on DNA , (2) the release of the completed

___—*Mw-‘In . ’a n i'll‘l
i’. ‘ ~
<\ . ‘

 

22

RNA chain from the ternary enzyme-DNA—RNA complex, and

(3) the release of enzyme from the ternary complex. There
are several kinds of termination which occur ig'yifigg
during RNA synthesis from double-stranded DNA, and one
kind is mediated by a protein factor called rho (p).

Nonspecific termination occurs at low ionic strength
(u:0.l) and is probably due to inhibition of enzyme by the
RNA product; termination occurs without release of either
enzyme or RNA(91). Specific termination is observed at
u=0.1 or higher with fd RF (45,73), ¢80 (46,73), T2 (93),
T4 (92—96), and T7 DNA (95,96), leading to the release of
RNA chains of discrete sizes which vary with the template.
The RNA chains synthesized from T4 and T7 DNA have pre—
dominantly uridine as the 3'—OH terminal nucleoside (95,96).
At u=0.11, RNA chains are released from T4 DNA but enzyme
does not reinitiate (94). At u=0.2 to 0.37, enzyme is
released and reinitiates RNA synthesis repeatedly (92,
94—96).

The Role of Rho Factor in Termination. A second
kind of specific termination requires the presence of the
termination factor rho (p) (97). This protein factor has
loeen extensively purified from E. gel; and is thought to
kxe a tetramer of molecular weight 200,000 (97). Rho-
induced termination is not very sensitive to either RNA

Exolymerase or DNA concentration, suggesting that rho is

 

 

 

 

23

not tightly bound at its receptor site (93). Rho does
not bind appreciably to DNA or to RNA polymerase, but does
bind to RNA (93). Rho is not a simple ribonuclease (93,

95). Termination in the presence of rho on T4, T A,

7!
$80, and fd RF DNA occurs at specific rho-dependent sites
which usually precede the rho-independent sites discussed

earlier (42,46,73,93,97). Rho is inactive at high ionic

 

 

"'7

strengths (u i 0.15) (93), and release of active enzyme

has not yet been observed after rho-induced termination

r. In' - ~"'.._’—‘ '1..'
‘1‘;

(92,93).

ll.

12.

l3.

14.

REFERENCES

Geiduschek, E. P., and Haselkorn, K., Ann. Rev. Biochem.,
2E, 647 (1969).

 

Richardson, J. P., Egogr. Nucl. Acid Res. M01. Biol.,
E, 75 (1969).

Yarus, M., Ann. Rev. Biochem., fig, 841 (1969)-

 

Geiduschek, E. P., Brody, E. N., and Wilson, D.,
Molecular Association in Biology, Academic Press,
New York, 1968, p. 163.

 

Chamberlin, M., gold Spring Harbor Symposia on Quanti-
tative Biology, 22, 851 (1970).

 

 

Burgess, R. R., Ann. Rev. Biochem., 3;, 711 (1971).

 

Silvestri, L. (Editor), First Lepetit Colloquium on
RNA Polymerase and Transcription, American
Elsevier Pfiblishing Company, Inc., New York,
1970.

 

 

Cold Spring Harbor Symposia on Quantitative Biology,
2E, Cold Spring Harbor Laboratory, New York, I970.

 

Burgess, R. R., J. Biol. Chem., 44, 6160 (1969).

*

 

Zillig, W., Zechel, K., and Halbwacks, H., Hoppe-Seyler's

 

Z. Physiol. Chem., 351, 221 (1970).

 

Berg, D., Barrett, K., and Chamberlin, M., in S. P.
Colowick and N. 0. Kaplan (Editors), Methods in
Enzymology, Vol. 21, Academic Press, New York,
p. 506.

 

 

Krakow, J. S., and Ochoa, S., Biochemische Z., 3 8,
796 (1963).

 

Krakow, J. S., and Fronk, E., J. Biol. Chem., 44,
5988 (1969).

 

Krakow, J. S., and von der Helm, K., Cold Spring Harbor
Eymposia on Quantitative Biolog , EE, 73 (1970).

24

 

:4
v”

 

15.
16.

17.

18.
19.
20.

21.

22.
23.

24.

25.

26.
27.
28.
29.
30.

31.

25

Johnson, J. C., DeBacker, M., and Boezi, J. A., J.
Biol. Chem., 246, 1222 (1971). ‘-

 

Burgess, R. R., Travers, A. A., Dunn, J. J., and
Bautz, E. F. K., Nature, 221, 43 (1969).

 

Burgess, R. R., J. Biol. Chem., 244, 6168 (1969).

 

Walter, G., Seifert, W., and Zillig, W., Biochem.
Biophys. Res. Commun., EE, 240 (19685.

Rudloff, R., and Zillig, W., as referenced by Burgess,
R. R., Egn. Rev. Biochem., 3E, 711 (1971).

 

Shapiro, A. L., Vinuela, E., and Maizel, J. V. Jr.,
Biochem. Biophys. Res. Commun., EE, 815 (1967).

Zillig, W., Fuchs, E., Palm, P., Rabussay, D., and .
Zechel, K., First Lepetit Colloquium on RNA ‘j
Polymerase and Transcription, 1970, p. 151. F

 

Berg, D., and Chamberlin, M., Biochemistry, 9, 5055
(1970). "

Travers, A. A., and Burgess, R. R., Nature, 222, 537
(1969).

 

Zillig, W., Zechel, K., Rabussay, D., Schachner, M.,
Sethi, V. 8., Palm, P., Heil, A., and Seifert, W.,
Cold Spring Harbor Symposia on Quantitative Biology,
EE, 47 (1970).

Richardson, J. P., J. Mol. Biol., EE, 83 (1966).

 

Stead, N. W., and Jones, 0. W., J. Mol. Biol., EE,
131 (1967).

Sternberger, N., and Stevens, A., Biochem. Biophys.
Res. Commun., E3, 937 (1966).

 

Fuchs, E., Millette, R. L., Zillig, W., and Walter, G.,
Eur. J. Biochem., E, 183 (1967).

Jones, 0. W., and Berg, P., J. Mol. Biol., EE, 199
(1966).

Stevens, A., Emery, A. J. Jr., and Sternberger, N.,
Biochem. BiOphys. Res. Commun., E3, 929 (1966).

Walter, G., Zillig, W., Palm, P., and Fuchs, E.,
Eur. J. Biochem., E, 194 (1967).

 

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

26

Pettijohn, D., and Kamiya, T., J. Mol. Biol., 29, 275
(1967). ‘—

 

Richardson, J. P., Proc. Natl. Acad. Sci. U.S.A., 55,
1616 (1966). ‘—

 

Crawford, L. V., Crawford, E. M., Richardson, J. P.,
and Slayter, H. S., J. Mol. Biol., E3, 593 (1965).

 

Freeman, E. J., and Jones, 0. W., Biochem. Biophys.
Res. Commun., EE, 45 (1967). re;

 

 

Bautz, E. K. F., and Bautz, F. A., Nature, 226, 1219
(1970).

 

 

l
Bautz, E. K. F., and Bautz, F. A., Cold Spring Harbor i
Eymposia on Quantitative Biology, 33, 227 (1970). j

 

Naono, S., and Tokuyama, K., Cold Spring Harbor Symposia %
or Quantitative Biology, 33, 375 (1970).

 

 

Summers, W. C., First Lepetit Colloquium on RNA Poly:
merase and Transcription, I970, p. 110.

 

 

Summers, W. C., and Siegel, R. B., Nature, 223, 1111
(1969).

 

Goff, G. G., and Minkley, E. G., First Lgpetit Colloquium
on RNA Polymerase and Transcription, 1970, p. 124.

 

 

Davis, R. W., and Hymen, R. W., Cold Spring Harbor
Eymposia on Quantitative Biology, 33, 269 (1970).

 

 

di Mauro, E., Snyder, L., Marino, P., Lamberti, A.,
Coppo, A., and Tocchini-Valentini, G. P., Nature,
222, 533 (1969).

Hinkle, D. C., and Chamberlin, M., Cold Spring Harbor
Eymposia on Quantitative Biology, 33, 65 71970).

 

 

Sugiura, M., Okamoto, T., and Takanami, M., Nature,
225, 598 (1970).

Okamoto, T., Suguira, M., and Takanami, M., Biochemistry,
E, 3533 (1970).

 

Szybalski, W., Kubinski, H., and Sheldrick, P., Cold
Epring Harbor Symposia on Quantitative Biology,
33, 123 (19667.

 

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

27

Summers, W. C., and Szybalski, W., Biochim. BiOphys.
Acta, 166, 371 (1968).

 

 

Sheldrick, P., and Szybalski, W., J. Mol. Biol., 29,
217 (1967). ‘—

 

Summers, W. C., and Szybalski, W., Virology, 34, 9
(1968). ‘—

Vogt, V., Nature, 223, 845 (1969).

 

Lancini, G., Pallanza, R., and Silvestri, L. G., 33 {—
Bact., El, 761 (1969). A

J. Antibiotics, 3;, 234 (1968).

 

Umezana, H., Mizuno, S., Yamazaki, H., and Nitta, K., i
I
‘l
l
l
l

Heil, A., and Zillig, W., FEBS Letters, El, 165 (1970).

 

Lill, H., Lill, U., Sippel, A., and Hartmann, G.,
First Leppetit Colloquium on RNA Polymerase and
Transcription, 1970, p. 55.

 

 

Sippel, A. E., and Hartman, G. R., Eur. J. Biochem.,
33, 152 (1970).

 

Goldthwait, P. A., Anthony, D. A., and Wu, C. W.,
First Lepgtit Colloquium on RNA Polymerase and
Transcription, 1970, p.710?

 

 

Sethi, V. S., Zillig, W., and Bauer, H., FEBS Letters,
E, 236 (1970).

 

Seifert, W., and Sethi, V. 8., Progress in Biophys.
Mol. Biol., E3, as quoted by Burgess, R. R.,
Ann. Rev. Biochem., 33, 711 (1971).

 

 

 

Darlix, J. L., Sentenac, A., Ruet, A., and Fromageot, P.,
Eur. J. Biochem., El, 43 (1969).

 

Maitra, U., and Hurwitz, J., Proc. Natl. Acad. Sci.
U.S.A., E3, 815 (1965).

 

Bremer, H., Konrad, M. W., Gaines, K., and Stent, G. S.,
J. Mol. Biol., 33, 540 (1965).

 

Maitra, U., Nakato, Y., and Hurwitz, J., J. Biol.
Chem., 242, 4908 (1967).

 

Bremer, H., and Bremer, R., Mol. Gen. Genet., 101,
6 (1968).

 

65.
66.
67.
68.
69.
70.
71.

72.
73.

74.
75.
76.

77.

78.

79.

80.

28

Anthony, D. D., Wu, C. W., and Goldthwait, D. A.,
Biochemistry, E, 246 (1969).

 

Wu, C. W., and Goldthwait, D. A., Biochemistry, E,
4450 (1969).

 

Wu, C. W., and Goldthwait, D. A., Biochemistry, E,
4458 (1969).

 

Downey, K. M., and So, A. G., Biochemistgzl 9, 2520
(1970). ‘

 

Furth, J. J., Hurwitz, J., and Anders, M., J. Biol.
Chem., 237, 2611 (1962).

Dunn, J. J., and Bautz, E. K. F., Biochem. Biophys.
Res. Commun., 33, 925 (1969).

 

 

Bautz, E. K. F., Bautz, F. A., and Dunn, J. J.,
Nature, 223, 1022 (1969).

Hall, B. 0., Fields, K., and Hager, G., First Lepetit
Colloquium on RNA Polymerase and Transcription,
1970, p. 148.

 

 

Takanami, M., Okamoto, T., and Suguira, M., Cold Spripg
Harbor Symposia on Quantitative Biology, 3E, 179
(19707 .

 

Okamoto, T., Sugiura, M., and Takanami, M., Biochemistry,
E, 3533 (1970).

 

Travers, A. A., Kamen, R. I., and Schleif, R. F.,
Nature, 228, 748 (1970).

 

Pettijohn,D. E., Clarkson, K., Kossman, C. R., and
Stonington, O. G., J. M01. Biol., EE, 281 (1970).

 

Cashel, M., 3. Biol. Chem., 44, 3133 (1969).

 

Cashel, M., Cold Spring Harbor Symposia on Quantitative
Biology, 3E, 407 (1970).

Travers, A., Kamen, R., and Cashel, M., Cold Spring
Harbor Symposia on Quantitative Biology, 3E,
415 (I9707.

 

Zubay, G., Schwartz, D., and Beckwith, J. P., Proc.
Natl. Acad. Sci. U.S.A., EE, 104 (1970).

 

81.

82.

83.
84.

85.

86.

87.
88.
89.
90.
91.
92.
93.
94.

95.

96.

97.

29

Emmer, M., De Crombrugghe, B., Pastan, I., and
Perlman, R., Epoc. Natl. Acad. Sci. U.S.A., 66,
480 (1970). _—'

 

Pastan, I. , and Perlman, R. L., Nature New Biology,
22 , 5 (1971).

 

Eron, L., Arditti, R., Zubay, G., Connaway, S., and
Beckwith, J. P., Proc. Natl. Acad. Sci. U.S.A.,
EE, 215 (1971).

 

 

 

F“
De Crombrugghe, B., Chen, B., Anderson, W. Nissley, P., l
Gottesman, M., Pastan, I., and Perlman, R., t
Nature New Biology, 231, 139 (1971). 1
a
Bremer, H., Cold Spring Harbor Symposia on Quantitative 3
Biology, 35, 109 (1970). 2
Richardson, J. P., J. Mol. Biol., g_, 115 (1966). p’

 

Sentenac, A., Simon, E. J., and Fromageot, P., Biochim.
BiOphys. Acta, 161, 299 (1965).

 

So, A. G., and Downey, K. M., as referenced by Burgess,
R. R., Ann. Rev. Biochem., 33, 711 (1971).

 

Pettijohn, D. E., Stonington, O. G., and Kossman, C. R.,
Nature, 228, 235 (1970).

 

Gariglio, P., Roodman, S. T., and Green, M. H., Biochem.
Biophys. Res. Commun., 31, 945 (1969).

 

Bremer, H., and Konrad, M. W., Proc. Natl. Acad. Sci.
U.S.A., E3, 801 (1964).

 

Maitra, U., and Barash, F., Proc. Natl. Acad. Sci.,
U.S.A., E3, 779 (1969).

 

Richardson, J. P., gold Spring Harbor Symposia on
Quantitative Biology, 3E, 127 (19703.

 

 

Millette, R. L., and Trotter, C. D., Proc. Natl.
Acad. Sci. U.S.A., EE, 701 (1970).

 

 

Millette, R. L., Trotter, C. C., Herrlich, P., and
Schweiger, M., Cold Spring Harbor Symposia on
Quantitative Biology, 3E, 135 (1970).

 

 

Maitra, U., Lockwood, A. H., Dubhoff, J. S., and Guha, A.,
Cold Spring Harbor Symposia on Quantitative Biology,
EE,'143 (1970).

Roberts, J. W., Nature, 24, 1168 (1969).

ART I CLE 1

2'-0-METHYLADENOSINE 5'-TRIPHOSPHATE. A SUBSTRATE FOR

DEOXYRIBONUCEIC ACID DEPENDENT RIBONUCLEIC ACID

POLYMERASE OF PSEUDOMONAS PUTIDA

BY

Gary F. Gerard, Fritz Rottman, and John A. Boezi

Reprinted from Biochemistry, EE, 1974 (1971)

 

3O

 

[Reprinted from Biochemistry, (1971 10, 1974.]
Copyright 1971 by the American Chemical Society and reprinte by permission of the copyright owner.

2’-0—Methyladenosine 5’-Triphosphate. A Substrate for
Deoxyribonucleic Acid Dependent Ribonucleic Acid Polymerase

of Pseudomonas putida'

Gary F. Gerard, Fritz Rottman, and John A. Boezi’r

ABSTRACT: 2’-0-Methyladenosine 5’-triphosphate (AmTP)
was a substrate for deoxyribonucleic acid dependent ribo-
nucleic acid polymerase of Pseudomonas putida. In the
DNA-directed synthesis of RNA with AmTP or an equirnolar
mixture of AmTP and adenosine 5’-triphosphate as the
adenyl substrates, a small amount of 2’-0-methyladenosine
5’-monophosphate (AmMP) was incorporated into RNA
after 60-min incubation. This amounted to 1—2 pmoles of
AmMP for each pmole of RNA polymerase added to the
reaction mixture. AmMP-containing RNA synthesized in
reaction mixtures which contained AmTP as the only added
adenyl substrate had a sedimentation coefficient of less than
4 S as determined by sucrose density gradient analysis.
Abortive release of the AmMP-containing RNA from the
DNA—RNA polymerase—nascent RNA ternary complex
did not occur. AmMP-containing RNA synthesized in
reaction mixtures which contained both AmTP and ATP

A contribution to the understanding of the mechanism of
the DNA-directed synthesis of RNA by RNA polymerase
(ribonucleoside triphosphate:RNA nucleotidyl transferase,
EC 2.7.7.6) has been made through studies of base-altered
analogs of the four common ribonucleoside triphosphates.
Knowledge of the extent and specificity of incorporation of
base-altered analogs by Escherichia coli RNA polymerase
has added to the understanding of the structural requirements
for hydrogen bonding between complementary bases in the
RNA synthetic process (Kahan and Hurwitz, 1962) and has
led to a definition of the role of base-stacking interactions
with nearest neighbors in RNA synthesis (Goldberg and
Rabinowitz, 1961; Nishimura et al., 1966; Slapikofl‘ and
Berg, 1967; Ikehara et al., 1968).

The mechanism of RNA synthesis has also been studied
through the use of substrate analogs altered in the ribose
moiety. Studies of 3'-deoxyadenosine 5'-triphoSphate (3 ’-
dATP) have indicated that a 3’-hydroxyl group is not re-
quired for binding of a nucleoside to E. coli or Micrococcus
luteus RNA polymerase or for incorporation of a nucleotide
into the 3’ end of RNA by these enzymes (Shigeura and Boxer,
1964; Shigeura and Gordon, 1965; Sentenac et al., 1968a,b).
Once 3’-dAMP is incorporated into the 3’ end of a RNA
chain, however, RNA chain growth is terminated. Another
substrate analog of ATP, 3’-amino-3’-deoxyadenosine 5’-

 

 

‘ From the Department of Biochemistry, Michigan State University,
East Lansing, Michigan 48823. Received December IO, 1970. This
work was supported in part by Grants 613-7543 and (313-7914 from
the National Science Foundation and by National Institute of Health
Training Grant GM 1091. Michigan Agriculture Experiment Station
Article No. 4743.

1' To whom to address correspondence.

1974 11,1971

BIOCHEMISTRY, VOL. 10, NO.

as the adenyl substrates was heterogeneous in size with most
of the RNA having a sedimentation coefficient of about 30 S.
Degradation of the RNA product by alkaline hydrolysis
followed by alkaline phosphatase digestion showed that 907,,
of the AmMP residues were located at the 3' end of the
RNA chain with the remaining 10% at the 5’ end or in the
interior of the chain. The studies with AmTP lead to the
following conclusions. (1) A free 2’-hydroxyl group is not
required for binding of an adenyl substrate by P. putida
RNA polymerase or for subsequent incorporation into
RNA. (2) Following the incorporation of AmMP into
the 3' end of the nascent RNA chain, the rate of RNA
chain growth is greatly reduced. The reduction in the rate
of RNA chain growth results in the accumulation of nascent
RNA chains with AmMP at the 3' end. The rate-limiting
step in RNA synthesis becomes the addition of the next
nucleotide to the nascent RNA chain.

triphosphate, appears to function in a similar manner as a
RNA chain growth terminator (Shigeura et a1., 1966).

Studies of analogs of nucleoside triphosphates altered at
the 2’ position of the ribose moiety have not clearly defined
the requirement for a 2’-hydroxyl group in reactions catalyzed ,
by RNA polymerase. Experiments using 2’-deoxyribonucleo—
side 5’-triphosphates have indicated that these analogs
probably do not bind appreciably to E. coli or M. Iuteus
RNA polymerase nor are they incorporated into RNA
in a native DNA-directed reaction (Weiss and Nakamoto,
1961 ; Chamberlin and Berg, 1962; Furth et 01., 1962; Anthony
et al., 1969). Other experiments using denatured calf thymus
DNA (Chamberlin and Berg, 1964) and poly[d(A-T)] (Krakow
and Ochoa, 1963) as templates indicated that the 2’-deoxy-
ribonucleoside 5’-triphosphates may have substrate activity.
Studies with 1-(B-D-arabinofuranosyl)cystosine 5'-triphos-
phate (ara-C'I'P)1 have shown that no appreciable DNA-
directed RNA synthesis by E. coli RNA polymerase occurred
when ara-CI‘ P was used in place of CTP (Cardeilhac and
Cohen, 1964). These studies were not designed, however,
to detect the direct incorporation of a small amount of
labeled ara-CMP into RNA. Chu and Fischer (1968) reported
that ara-C is incorporated into both terminal and internal
positions of RNA of murine leukemic cells in viva indicating
that for this system an unaltered 2’ position on the ribose
moiety of a nucleoside triphosphate is not required for
incorporation into RNA.

This report describes the effects of 2’-0-methyladenosine
5’-triphosphate (AmTP) on the reactions catalyzed by RNA

 

1The abbreviation used that is not listed in Biochemistry 9. 4022
(1970), is: ara-C, 1-(fi-D—arabinofuranosyl)cytosine.

 

l

2'-0-METHYLADENOSINE 5 '-TRIPHOSPHATE

polymerase of Pseudomonas pulida. The purpose of this
investigation was twofold. First, we wished to establish
whether or not AmTP was a substrate for RNA polymerase,
and if so, what effect the incorporation of AmMP into the
3’ end of nascent RNA would have on further RNA chain
growth. Second, we hoped to prepare 2’-0-methyl-containing
RNA which was complementary to a DNA template. 2’-0-
Methylribonucleoside diphosphates have been used to prepare
various synthetic polymers using polynucleotide phos-
phorylase (Rottman and Heinlein, 1968; Rottman and
Johnson, 1969). The results of the present studies demon-
strated that AmTP is a substrate for P. putida RNA polym-
erase. Following the incorporation of AmMP into RNA,
the rate of RNA chain growth was greatly reduced. Conse-
quently, the synthesis of RNA chains containing more
than a few AmMP residues was infeasible.

Materials and Methods

Materials. NADP, UDPG, glucose 6-ph05phate dehydrog-
enase, and all the unlabeled 5’-phosphate derivatives of the
ribonucleosides were purchased from P-L Biochemicals, Inc.
Calf thymus DNA, herring sperm DNA, poly(A), poly(U),
poly(C), and phosphoglucomutase were obtained from
Sigma Chemical Co. Poly[d(A-T)] was from Miles Labora-
tories, Inc. E. coli alkaline phosphatase, electrophoretically
purified, was from Worthington Biochemical Corp. [’H]ATP
and [’H]adenosine were obtained from Schwarz BioRescarch,
Inc. Whatman DEAE-cellulose (DE-1) was from Reeve
Angel. Nitrocellulose membrane filters (type B-6) were
obtained from Schleicher & Schuell. Wheat germ RNA
(Singh and Lane, 1964), E. coli B tRNA (Holley et al., 1961),
and Pseudomonas putida rRNA (Payne and Boezi, 1970)
were prepared by published procedures. P. putida bacterio-
phage gh-l DNA (Lee and Boezi, 1966, 1967) was prepared
by the method of Thomas and Abelson (1966). P. putida
RNA polymerase was purified by the procedure of Johnson
et al. (1971). The preparations of RNA polymerase used in
these experiments were at least 90% pure. UDPG pyro-
phosphorylase was isolated from calf liver (Albrecht et al.,
1966) and recrystallized twice.

Descending paper chromatography employed one of the
following solvent systems: (a) isobutyric acid-concentrated
NH40H—H20 (66:1 :33, v/v), (b) isopropyl alcohol—con-
centrated NH40H~H20 (7:1 :2, v/v), and (C) iSOPFOPyl
alcohol—concentrated NH.OH—0.l M boric acid (7:1:2,
v/v).

AmTP was prepared by the procedure of Rottman and
Heinlein (1968). AmTP was further purified by descending
Chromatography on acid-washed Whatman No. 3MM
paper. Successive development of AmTP was carried out
in solvent systems a and b, and in each case, development
was followed by elution from the paper and lyophilization
to concentrate the eluant. The purified AmTP contained a
small amount of AmDP (about 5%) as determined by chro-
matography in solvent 3.

The synthesis of [3H]AmTP from [3H]adenosine was
carried out by methods similar to those described by Rottman
and Heinlein (1968). The following modifications in procedure
were made. (1) [3H]AmMP prepared from [3H]Am and
UMP with wheat seedling phosphotransfcrase was purified
by descending chromatography on Whatman No. 3MM
paper 1n solvent a, and (2) [3H]AmTP prepared from [3H]-
AmMP and ATP by treatment with rabbit muscle nD/Oklnuse
W35 Purified in two steps. First, [iHlAmTP and [3H]AmDP

were separated from ['HJAmMP and the 5’-phOSPha"e
derivatives of adenosine by descending chromatography
in solvent c. Second, [‘H]AmTP was separated from PH)-
AmDP by electrophoresis in 0.05 M ammonium formal!C
(pH 3.5) on acid-washed Whatman No. 3MM paper US$13 a
Hormuth pherograph (Brinkman Instruments, Inc.) (1800 V°
2 hr). The purified [3H]AmTP contained a small amount
of [‘H]AmDP (about 10%) as determined by chromatogl'aph.y
in solvent a.

Analytical Methods. Protein concentrations were deter-
mined by the method of Lowry et a1. (1951), using boVInC
serum albumin as a standard. The concentrations of 113““
gh-l and calf thymus DNA were determined speCtFOPho‘
tomctrically based on the extinction coefficient £263: 200'
The molar extinctions, e(P), used to determine the homopoh"
mer concentrations were: 10.5 X 103 at 257 nm, 9.2 X If);
at 260 nm, and 6.5 X 103 at 267 nm for poly(A), poly(U).
and poly(C), respectively, in 0.1 M NaCl—0.05 M Tris-acetate:
(pH 7.5)(1'5‘0 et al., 1962). For poly[d(A-T)], .(P) = 6-7 A
103 at 260 nm and pH 7.5 was used (Radding and Kornberg.
1962)

Assay of RNA Polymerase. The reaction miXture
monitoring the gh-l DNA-directed synthesis of RNA ‘3!
radioisotope incorporation contained 20 mM Tris—acetate
(DH 8-0), 5 mM 2-mcrcaptocthanol, 2 mM magnesium acetzuf-
0.5 mM manganese acetate, 60 mM ammonium 35‘3““;
0.2 mM ATP, and/or 0.2 mM AmTP, 0.2 mM each of UT];
GTP, and CTP, 100 pg/ml of gh-l DNA, and P. putid" 3.”.
POIYmerase. The concentrations of divalent metal ‘0;
(Mg2+ and Mn“), monovalent cation (NHP‘). nucleony;
triphosphates, and DNA used in the reaction mi?mire gags
the optimal rate of RNA synthesis. ATP or AmTP W'

.’ >d a
labeled with 3H. RNA polymerase (aafifi'a) contzlrtmc’
n -

for

full compliment of a' factor. After incubatio Smiied
the total reaction mixture or a sample from 1‘ ‘22:)” 10“;
with 100 ptl of 0.1% sodium dodecyl sulfate.te 30mm“

trichloroacetic acid—l 7", sodium pyrophospha 1 addcd'
(5 ml) and 250 pg of herring sperm DNA were t“.mcgllt’t‘ti‘l
After 15 min at 0—4°, the acid-insoluble Produd was C"1h four
on a membrane filter. Each filter was washed 07 Sodium
S-ml portions of cold 107o trichloroz‘lCetic ac‘d’q/ilimctiuh
Pyrophosphate, dried, and then monitOred {or r‘gcimillaion
using liquid scintillation spectrometry' _buwlbcnzofl‘
fluid (5 ml) contained 4 g of 2,5 bisLZ—(S-‘e’t ' u
zolyl)]thiophene/l. of toluene. 0N "‘
Reactions using the templates Poly[d(AfsnptlA “i”
poly(A), poly(C), and denatured calf Ehym (10““5031
monitored using a spectrophotort16
et al., 1968). The assay couples the
pyrophOSphate from ribonucleoside
erization to NADP reduction. The: Eta“: p .. W"
poly[d(A-T)] contained 20 mM Tris-QCLTP an 019- "

C Yeti"

manganese acetate, 04 mM each of dust- . g

poly[d(A-T)]. and 27 “gm! of RNA r) 0 1 3"
mixture with poly(U) contained 10()

8.0), 1 mM manganese acetate, 0.4 ml‘r1 «1‘6
and 20 pg ml of RNA polymerase, ’ 3 ,ztccl“
with poly(A) contained 20 mM Tr’ "1;" 36 #1
mM manganese acetate, 2.8 mM UT I ” cl ‘
20 ,ugr'ml of RNA polymerase. The 3" ts»
poly(C) contained 100 [TIM Tris-ace 1; k #5
manganese acetate, 1.2 mM GTP, 28’ ‘- ?‘
pg’ml of RNA polymerase. The react 1) ’
turcd calf thymus DNA contained 2 ‘- ‘ ~ g
8.0), 2 mm magnesium acetate, 0.5 r1 I, ‘-

nunrurzurQTuv \Jnl 10 run.

. "109

 

Ma». “-3“;-

 

L7: 7' «f'

iii.

if? ’

0.4 mM ATP, 94 pg/ml of denatured calf thymus DNA,
and 93 pig/ml of RNA polymerase. In addition, all of the
above reaction mixtures contained 0.4 mM NADP, 0.4 mM
UDPG, and excess phosphoglucomutase, glucose 6-phosphate
dehydrogenase, and UDPG pyrophosphorylase. For the
concentration of divalent metal ion used in each type of
reaction, the optimal concentrations of nucleoside triphos-
phate(s) and template were used. The amount of RNA
polymerase used in each case was rate limiting.

Analysis of the Positions of the [’H]AmMP Residues in RNA.
Reaction mixtures (0.5 ml) containing [3H]AmTP (97,000
cpm/nmole), or [3H]AmTP and ATP, 95 pg/ml of RNA
polymerase, and the other components of the reaction mixture
for the gh-l DNA-directed synthesis of RNA listed above

were incubated for 60 min. After incubation, a sample of

each reaction mixture was removed and assayed for 3H-
labeled trichloroacetic acid insoluble product. The remainder
of each reaction mixture was then treated in the following
manner. Trichloroacetic acid to 5% and 2 mg of wheat
germ RNA were added. The resultant precipitate was collected
by centrifugation and washed by resuspension in five 5-ml
portions of cold 5% trichloroacetic acid, once with 5 ml of
95% ethanol, and once with 5 ml of ethanol—ether (3:1,
v/v). The washed precipitate together with an additional
8 mg of wheat germ RNA was dissolved in 1 ml of 1.0 N
KOH and incubated at room temperature for 90 hr. Cold
4 N HC104 (0.25 ml) was added and the mixture was centri-
fuged at 0°. The pellet was washed twice with 1 ml of cold
0.2 N HClO4. The supernatant from the first centrifugation
and the washings were combined, neutralized with KOH,
and centrifuged to remove KClO4. The neutralized solution
which contained 200 A250 units was diluted to 6.0 ml and
made 0.05 M in (NH4)2C03 (pH 8.5). Following the addition
of 0.3 mg of E. coli alkaline phosphatase, the solution was
incubated at 37° for 24 hr. After lyophilization, the phos-
phatase digest was dissolved in 1 ml of H20. Samples were
removed to assay radioactivity by liquid scintillation spec-
trometry (Bray, 1960). The rest of the phosphatase digest
was diluted to 100 ml and applied to a DEAE-cellulose
column (1.2 x 15 cm) in the formate form at a flow rate of
1 ml/min. The nucleosides were eluted from the column with
0.005 M ammonium formate followed by the dinucleoside
monophosphates with 0.5 M ammonium formate (Price and
Rottman, 1970). After repeated lyophilization, samples from
each fraction were monitored for radioactivity and for
absorbancy at 260 nm. Based on the absorbancy at 260 nm

v of the nucleoside and dinucleoside monophosphate fractions,

the 2’-0-methyl content of the carrier wheat germ RNA

was 2.5%. The value reported by Singh and Lane (1964)

for wheat germ RNA is 1.9%. The total recovery of A250
units from the DEAE-cellulose column for each type of
reaction mixture was 97% of the Am units present in the

* neutralized KOH hydrolysate.

Identification of the 3H-labeled material in each fraction

. was carried out in the following manner. The nucleoside

5. fraction was chromatographed on acid-washed Whatman
; . No. 3MM paper in solvent a. This procedure separated

.. A and Am from G, U, and C. A and Am were then separated

by chromatography in solvent c. The dinucleoside mono-
Phosphate fraction was subjected to electrophoresis on
Whatman No. 1 paper in 0.1 M ammonium bicarbonate

- (pH 7.8) using a Spinco Model R instrument (400 V, 4 hr).
' . Separation of dinucleoside monOphosphates (AmpN) from

any contaminating nucleosides is achieved by this procedure.
Sucrose Density Gradient Analysis of [3H]AmMP-Con-

Iglh nrnrunulc'rnv \Inl 10 run 11 1071

GERARD, ROTTMAN, AND BOEZI

taining RNA and the gh-I DNA-(RNA Polymerase)-(Nascent
RNA) Complex. Reaction mixtures (0.25 ml) containing
[3H]AmTP (70,000 cpm/nmole), or [3H]AmTP and ATP,
95 pg/ml of RNA polymerase, and the other components
of the reaction mixture for the gh-l DNA-directed synthesis
of RNA as listed above were incubated for 60 min. After
cooling to 4°, a sample (0.1 ml) was removed from each
reaction mixture and mixed with 10 ul of 5% sodium dodecyl
sulfate. The mixture was incubated at -37° for 5 min. After
cooling to 4°, the mixture was centrifuged at 5000 rpm
for 5 min to remove precipitated sodium dodecyl sulfate.
Samples (0.1 m1) of the original reaction mixtures and
the sodium dodecyl sulfate treated samples were layered
on 4.8-ml 5—20% linear sucrose gradients prepared in 50
mM Tris-acetate (pH 8.0), 100 mM NaCl, and 1 mM dithio-
thrietol. After centrifugation for 260 min at 39,000 rpm
in the Spinco SW39 rotor at 4°, fractions were collected
from the bottom of the centrifuge tubes. To determine the
position of [3H]AmMP-containing RNA in the gradients,
each fraction or a sample of each was precipitated with cold
10 % trichloroacetic acid-1 % sodium perphosphate, collected
on a membrane filter, and monitored for radioactivity.

The position of DNA in the gradients from the original
reaction mixtures was determined by assaying samples
(50 ,ul) of each fraction for template activity, measured as
the incorporation of [“C]AMP into RNA in the presence of
excess RNA polymerase. At 32 tug/ml of RNA polymerase,
the rate of [“C]AMP incorporation into RNA was prOpor-
tional to the DNA concentration between 0 and 3 pg per ml.
In the sucrose density gradient analyses, E. coli B tRNA,
P. putida rRNA, and gh-l DNA were used as markers of
known sedimentation coefficients.

Results

Incorporation of [3HlAmMP into RNA. [3H]AmMP was
incorporated into RNA in a gh-l DNA-directed reaction
catalyzed by Pseudomonas putida RNA polymerase (Table
I). In the two reaction mixtures which contained RNA
polymerase, gh-1 DNA, GTP, CT P, UTP, and [’H]AmTP,
55 and 66 pmoles of [3H]AmMP per ml were incorporated
into RNA. If ATP at a concentration equal to that of AmTP
was included in the reaction mixture, approximately the
same amount of [3H]AmMP was incorporated.

Relative to the amount of AMP incorporated into RNA,
the amount of AmMP incorporated was small (Table I).
In comparable reaction mixtures, one containing [’H]ATP,
the other [3H]AmTP, approximately 200 times as much
[3HJAMP (11,400 pmoles/ml) was incorporated into RNA
as [3H]AmMP (55-66 pmoles/ml). In the reaction mixtures
containing an equimolar mixture of ATP and AmTP, 77
times as much [3H]AMP (5800 pmoles/ml) as [3H]AmMP
(75 pmoles/ml) was incorporated into RNA.

Sucrose Density Gradient Analysis of [3H'JAmMP-Con-
taining RNA. The sedimentation patterns of [’H]AmMP-
containing RNA synthesized in gh-l DNA-directed reactions
with [3H]AmTP or an equimolar mixture of [3H]AmTP
and ATP as the adenyl substrates were examined by sucrose
gradient centrifugation. The results are presented in Figure 1.
Most of the [3H]AmMP-containing RNA synthesized in the
reaction mixture which contained [3H]AmTP as the adenyl
substrate was small in size, as judged from a sedimentation
coefficient of less than 4 S (Figure 1, lower). [’H]AmMP-
containing RNA synthesized in the reaction mixture which
contained both [3H]AmTP and ATP was of diverse size,

m3“ "’ 7 '~‘.

2 '-0-METHYLADENOSINE 5 ’-rnrruosenxre

 

TABLE I: Incorporation of [-‘HlAmMP or [‘H]AMP into RN A.°

 

 

 

Adenyl Substrate(s) [‘HlAmMP or
Added to the Reaction Incubn Time [’HlAMP Incorpd
Expt Mix. Reaction Mix. Component Omitted (min) (pmoles/ml)
1 [3H]AmTP None 0 0
[3H]AmTP None 30 66
[’HlAmTP RNA polymerase 30 0
[3H]AmTP, None 30 75
ATP
[’H]AmTP, gh-l DNA 30 4
ATP
AmTP, ATP None: [3H]AmTP 30 0
added immediately
before reaction was
terminated
2 ['HlAmTP None 30 55
['HlAmTP, None 30 72
ATP
3 [’I-I]ATP None 0 0
WHAT? None 30 1 1 , 400
[3H]ATP gh-l DNA 30 57
[3H]ATP, None 30 5 , 800
AmTP

4.—

0 Reaction mixtures (0.05 ml) containing the components described in Materials and Methods for the gh-l DNA-directed
synthesis of RNA were incubated for the times indicated. After incubation, the amount of radioactivity in the trichloroacetic acid
insoluble product was determined as described in Materials and Methods. [3H]AmTP and [3H]ATP where indicated were present
at 70,000 and 15,000 cpm per nmole, respectively. RNA polymerase was at 63 tug/ml. The values for the incorporation of [3H]-
AmMP and [3H]AMP were those obtained after the subtraction of 16 and 35 pmoles per ml, respectively, which were the amounts
of 3H-labeled substrates which were trapped on the membrane filters from reaction mixtures terminated at zero time.

!__-

TABLE 11: Position of the [3H]AmMP Residues in RNA.°

 

 

C13CCOOH-
Insoluble RNA Alkaline Dinucleoside
Adenyl Substrate(s) Added Product Formed Phosphatase Nucleoside Fraction Monophospham
Expt to the Reaction (cpm) Digest (cpm) (cpm) Fraction (cpm)
If,
1 [3H]AmTP 1 1 , 000 500
2 [3H]AmTP, ATP 10,000 9700 7700 900
//

 

4' Reaction mixtures (0.5 ml) containing the components for the gh-l DNA-directed synthesis of RNA described in Malt”ls

and Methods were incubated for 60 min, and were treated as described in Materials and Methods.
/

 

with most of the [3H]AmMP-containing RNA having a For the analysis of the RNA synthesized in the reacIIO"
sedimentation coefficient of about 30 S (Figure 1, upper). mixture which contained [3H]AmTP as the adenyl subSlfiltfi

Position of the [3H]AmMP Residues in RNA. The RNA only 5% of the radioactivity in the [3H]AmMP-C0m‘““m%
produced in gh-l DNA-directed reactions with [3H]AmTP RNA product was recovered after the washing Prowl”L
or [3H]AmTP and ATP as the adenyl substrates was washed with trichloroacetic acid prior to KOH hydrolysis. AS Sh?“ |
with cold trichloroacetic acid to remove unreacted substrates by the sucrose density gradient analysis (Figure 13 10W 'l'
and was then degraded with KOH followed by alkaline [3H]AmMP-containing RNA produced in the reactio n
phosphatase. As a result of this procedure, AmMP residues ture with [3H]AmTP as the adenyl substrate was 5““ 1

n le’

 

at the 3’ end of the RNA chain were converted into Am, size. Consequently, it was lost during the repealedfasmfi
and those at the 5’ end or in the interior of the RNA chain with cold trichloroacetic acid. Insufficient radioacmm: M)
were converted into dinucleoside monOphosphates (AmpN). present in the alkaline phosphatase digest (Table “i “$110-
Separation of Am and AmpN was achieved by chromatog- to permit analysis of the nucleoside and dint)CleOSidc mo“
raphy on DEAE-cellulose. The results are shown in Table II. phosphate content by DEAE-cellulose chromatography"

107.7

 

asasé

 

CPU MANN? INCORPORATED

o s§§§§

 

 

 

IO 20 3O
FRACTION W

, nouns l: Sucrose density gradient analysis of the ['H]AmMP—

15 containing RNA product. Reaction mixtures containing the com-
'. ponents of the gh-l DNA-directed synthesis of RNA with [’11]-

AmTP (lower) or [3H1AmTP and ATP (upper) were treated with
sodium dodecyl sulfate and analyzed by sucrose gradient centrifuga-
tion as described in Materials and Methods.

For the analysis of the RNA synthesized in the reaction

/

/
t7 . g 3
ti " K g El
iis- S g
_.. 2 §
0. 2
g g
3 a
/ 0

mixture which contained [’HJAmTP and ATP as the adenyl

-’ substrates (Table II, expt 2), 97% of the radioactivity in the

" ['HlAmMP-containing RNA product was recovered in

7 ' the alkaline phosphatase digest. The alkaline phosphatase
5"" j digest was separated into nucleoside and dinucleoside mono-
phosphate fractions by chromatography on DEAE-cellulose.
3‘ The nucleoside fraction contained 90% of the radioactivity

recovered from the DEAEocellulose column and the dinucleo-
side monophosphate contained 10%. Analysis of the nucleo-

 

 

 

 

 

 

FRACTION ”BER

‘v FIGURE 2: Sucrose density gradient analysis of the DNA-(RNA
cf: P0|ymerase)-([3H]AmMP-containing RNA)

ternary complex.

Samples of reaction mixtures containing the components of the gh-l

" DNA-directed synthesis of RNA with [3H]AmTP (lower) or [all]-
r’: AmTP and ATP (upper) were analyzed by sucrose gradient
1 centrifugation or treated with sodium dodecyl sulfate and then

3;: " allailed by sucrose gradient centrifugation. (0) [3H]AmMP-con-
, taming RNA of the samples that had been treated with sodium dode-
" ., Cyl sulfate. (A) [3H1AmMP-containing RNA of samples that had not

it treated with the detergent. (0) DNA, as monitored by template

attivity measured as the incorporation of [“C]AMP into RNA in

.r.,;. the presence of an excess of RNA polymerase.

*h’ 1972 nvnnuusnrc'rnv uni In turn 11

1071

 

 

 

 

GERARD, ROTTMAN, AND 3032]

E I l l l I T
B
u 4°"
73'
S ”r '
0
3
seat .
i
0, IO ~
Ill
2’
c l l l L l l

O 40 80 IZO ISO 200 240

TIME (MINUTE9

FIGURE 3: Effect of AmTP on the incorporation of ['l-flAMP into
RNA. The reaction mixtures contained, in a volume of 0.5 ml, those
components for the gh—l DNA-directed synthesis of RNA listed
in Materials and Methods. RNA polymerase was present at 33
pg/ml and ['HlATP at 15,000 cpm/nmole. At various times during
the incubation, SO-pl aliquots were removed and assayed for radio-
activity in the trichloroacetic acid insoluble product as described in
Materials and Methods. Incorporation of ['l-llAMP is shown in the
absence of AmTP (O), and in the presence of 0.2 mM (A) and 0.8 man
(I) AmTP.

side fraction from the DEAR-cellulose column by paper
chromatography confirmed that the radioactive material
was [’H]Am. Analysis of the dinucleoside monophosphate
fraction by paper electrophoresis showed that the fraction
was not contaminated with labeled Am and that the radio-
active material present had the same electrophoretic mobility
as authentic dinucleoside monophosphates.

Nome/ease of [’MAmMP—Containing RNA from the gh-I
DNA - (RNA Polymerase) - (Nascent RNA) Complex. The
data presented in Table II led to the conclusion that most of
the [’HJAmMP incorporated into RNA was at the 3’ end
of RNA chains. A possible explanation for this conclusion
is that the addition of an AmMP residue to the 3’ end of a
nascent RNA chain induced the release of that chain from
the DNA-(RNA polymerase)'(nascent RNA) ternary com-
plex before RNA chain propagation could continue. To
look for release of [’H]AmMP-containing RNA, sucrose
density gradient analysis was used. After incubation, samples
of reaction mixtures containing the components of gh-l
DNA-directed RNA synthesis with [3H]AmTP or [3H]AmTP
and ATP as the adenyl substrates were treated with sodium
dodecyl sulfate, and centrifuged through sucrose gradients.
The position in the gradients of [3H1AmMP-containing RNA,
that is [3H]AmMP-containing RNA released from the ternary
complex, was determined. Other samples of the reaction
mixtures, not treated with sodium dodecyl sulfate, were also
centrifuged through sucrose gradients. The position in the
gradients of the DNA - (RNA polymerase)-(nascent RNA)
ternary complex was determined to be that position where
[3H]AmMP-containing RNA and DNA cosedimented at
speeds faster than that of free [3H]AmMP-containing RNA
or free DNA. The results are presented in Figure 2. In the
case of the analysis of the reaction mixture containing [3H]-
AmTP as the adenyl substrate, the data show that approx-
imately 85% of the [3H]AmMP-containing RNA cosedi-
mented with DNA in the ternary complex. This result leads
to the conclusion that release of [3H]AmMP-containing RNA
from the ternary complex had not occurred to any significant
degree. In the case of the analysis of the reaction mixture
containing both [3H]AmTP and ATP as the adenyl substrates,
it is not clear how much, if any, release of [3H]AmMP-

 

2'-0-METHYLADBNOSINE 5 '-TRIPHOSPHATE

 

TABLE III: The Effect of AmTP on Several Reactions Catalyzed by RNA Polymerase.“

 

Initial Velocity:
nmoles of NADPH

 

Expt NTP Substrates Added Polymer Added AmTP (mM) Produced/min per ml Inhibn (7,)
1 ATP, UTP Poly[d(A-T)] 2 .6
ATP, UTP Poly[d(A-T)] 0 .4 l .5 42
UTP Poly[d(A-T)] 0 . 1
UTP Poly[d(A-T)] 0 .4 0 . 1
ATP, UTP <0 .1
2 ATP Poly(U) l .0
ATP Poly(U) 0 .4 0 . 5 50
Poly(U) 0 .4 0
ATP 0
3 ATP dDNAb l .4
ATP dDNAb 0.4 0.2 86
dDNAb 0 .4 0
ATP 0
4 UTP Poly(A) 3 .1
UTP Poly(A) 0.2 3 .1 0
UTP Poly(A) l .0 3. 0
Poly(A) 2.8 0
UTP 0
5 GTP Poly(C) 4 . 5
GTP Poly(C) 0 .4 4 .2 7
Poly(C) 0.4 0
GTP 0

 

° The reaction mixtures, in a volume of 0.25 ml, contained the components described in Materials and Methods. The production
of NADPH was measured as described in Materials and Methods. 5 dDNA-denatured calf thymus DNA.

containing RNA from the ternary complex had occurred
because of the difficulty in distinguishing between free RNA
and RNA in the ternary complex.

Absence of Synthesis of Trichloroacetic Acid Soluble
Oligonucleotides. In all of the experiments described above
in which AmTP was the adenyl substrate, only the formation
of trichloroacetic acid insoluble RNA was monitored. In
order to determine if trichloroacetic acid soluble oligonucleo-
tides were being synthesized, the reaction was monitored by
measuring inorganic pyrophosphate formation using a
spectrophotometric assay (Materials and Methods). In a
reaction mixture containing 50 pg/ml of RNA polymerase,
100 rig/ml of gh-l DNA, 0.2 mM each of GTP, CT P, UTP,
and AmTP, and the components of the coupled assay system,
no inorganic perphosphate formation was observed. Thus,
within the limits of sensitivity of the assay (a rate of formation
of 0.02 nmole of inorganic pyrophosphate/min per ml of
reaction mixture or greater), no DNA-directed reaction was
detected, indicating that RNA polymerase was not repeatedly
initiating the synthesis and release of trichloroacetic acid
soluble oligonucleotides.

AmTP as an Inhibitor of RNA Polymerase Catalyzed
Reactions. Although the presence of ATP at an equimolar
concentration to that of [3H]AmTP did not decrease the
amount of [3H]AmMP incorporated into RNA (see Table l),
the presence of AmTP in reaction mixtures containing
[3H]ATP decreased the amount of [3H]AMP incorporated.
In reaction mixtures containing RNA polymerase, gh-l
DNA, and 0.2 mM each of GTP, CT P, UTP, and [3H1ATP,
the initial rate of [3H]AMP incorporation into RNA was

!__

inhibited by 63 and 90% in the presence of 0.2 and 0.8 rmt
AmTP, respectively (Figure 3). The extent of PHIAMP
incorporation at 240 min was reduced by 30 and 7292-
respectively, by the two concentrations of AmTP.

The polymerization of ATP and UTP directed by POI)"
[d(A-T)], and the polymerization of ATP directed by pONL'
or denatured calf thymus DNA were inhibited by AmTP
(Table III). AmTP had little or no effect on the 1395‘“;
directed polymerization of UTP or the poly(C)-dlra‘tct-
polymerization of GTP. These results indicate that AmTP
was an inhibitor of only those reactions which used A“)
as a substrate.

Discussion

AmTP was a substrate for the gh-l DNA-directed syn’h"5"
of RNA by Pseudomonas putida RNA polymer?“ "‘
virtue of its structure and the fact that it inhibited only ‘th
RNA polymerase catalyzed reactions which used ATP as?
substrate, AmTP may be regarded as a substratecans‘lf:E
of ATP. Thus, a free 2’-hydroxyl group is not fequ'red hf
binding of an adenyl substrate by P. putida RNA Po‘ymebrig
or for subsequent incorporation into RNA. Howevcm“ M
gh-l DNA-directed reactions studied, the amOum 0‘ M6”,
incorporated was small. With either AmTP of an equtmro"
mixture of AmTP and ATP as the adenyl subsifa‘f’s‘ 8.53m“;
imately 1—2 pmoles of AmMP was incorporated in (low,
pmole of enzyme added to the reaction mixtureS- 69210“.
RNA polymerase was calculated using mol wt5 X ‘0
son et al., 1971).

 

 

. /.

 

Q
n .............. nnl In sun 11 1971 1m

h

\

5r

AmMP-containing RNA synthesized in the reaction mixture
which contained AmTP as the only added adenyl substrate
was small in size, as judged from a sedimentation coefficient
of less than 4 S. AmMP-containing RNA synthesized in the
reaction mixture which contained AmTP and ATP was
heterogeneous in size with most of the RNA having a sedi-

‘ mentation coefficient of 30 S. Under similar assay conditions,

RNA synthesized in gh-l DNA-directed reactions with the

" four common nucleoside triphosphates (GTP, CT P, UTP,

and ATP) can achieve an even larger size, as judged from a
sedimentation coefficient of 45 S (G. F. Gerard and J. A.
Boezi, unpublished results).

AmMP incorporated into RNA in the reaction mixture

‘ which contained an equimolar mixture of AmTP and ATP

was found both at the 3’ end of the RNA chain, and at the
5’ end or in the interior of the chain. Most of the AmMP
was at the 3’ end of the chain. The presence of some AmMP
at the 5' end or in the interior of the RNA chain shows that
nascent RNA chain growth can continue following the
incorporation of AmMP. Thus, AmTP does not function
as a chain terminator as is the case with 3’-dATP (Shigeura
and Boxer, 1964).

The location of AmMP incorporated into RNA in the
reaction mixture which contained AmTP as the only added
adenyl substrate was not determined. Since there was no
significant RNA chain release from the ternary complex,
each enzyme molecule could have initiated the synthesis
of only one RNA chain. Because there was only 1—2 pmoles
of AmMP incorporated per pmole of enzyme, there was there-
fore probably only 1—2 AmMP residues/RNA chain. Most
of the AmMP-containing RNA synthesized in this reaction

. mixture was estimated to have a sedimentation coefficient of

somewhat less than 4 S. As estimated from the sucrose
gradient profile, the sedimentation coefficient was approx-
imately 3 S which corresponds to a chain length of 50—75
nucleotides (Madison, 1968). In order to explain the synthesis
of RNA of this chain length containing only 1—2 AmMP
residues, two alternatives are suggested. First, the RNA had
an unusual base composition being very low in adenyl
content. Second, the RNA had a base composition which
reflected the overall base composition of the gh-l DNA
template (Lee and Boezi, 1966) with a mole per cent adenyl
groups equal to 22%, but the additional adenyl groups were
from ATP present as a contaminant in the reaction mixtures.
Consistent with the second alternative is the observation that
a slow rate of RNA synthesis (approximately 0.5% of the
control rate) was detected in reaction mixtures which con-
tained CTP, GTP, and UTP, but no added ATP (G. F.
Gerard and J. A. Boezi, unpublished results). This observation
suggests that one of the stock nucleoside triphosphates was
contaminated with a small amount of ATP.

The results presented in this paper lead us to the following
understanding of the mechanism of incorporation of AmMP
into RNA by P. putida RNA polymerase. The substrate,
AmTP, binds on the enzyme surface in place of ATP, and
is incorporated into the 3’ end of the nascent RNA chain.
The rate of incorporation of AmMP into the 3’ end of RNA
is probably slower than that of AMP. This reduced rate is due
to a lower affinity of the enzyme for AmTP than for ATP
andg'or a slower rate of phosphodiester-bond formation with
AmTP as substrate than with ATP. The rate of initiation of
synthesis of RNA chains is probably also slower with AmTP
than with ATP. If ATP is present in the reaction mixture,
competition between AmTP and ATP for binding on the
enzyme results. Following the incorporation of AmMP

A---

GERARD, ROTTMAN, AND 8032!

into the 3’ end of the nascent RNA chain, the rate of RNA

. chain growth is greatly reduced. The reduction in rate of RNA

chain growth results in the accumulation of nascent RNA
chains with the AmMP at the 3’ end. The bulky 2'-0-methyl
group at the 3’ end of the nascent RNA chain could impede
translocation of RNA polymerase relative to the DNA
template and nascent RNA chain and/or could hinder
phosphodiester-bond formation with the incoming nucleoside
triphosphate. Whatever the reason, the addition of the
next nucleotide to the chain becomes the rate-limiting step
in RNA synthesis. Abortive release of the RNA from the
ternary complex does not occur. Following the addition of the
next nucleotide, the rate of RNA chain growth may return
to the normal in vitro rate, but perhaps not until the AmMP
residue in the nascent RNA chain leaves the enzyme surface.

References

Albrecht, G. J., Bass, S. T., Seifert, L. L., and Hansen, R. G.
(1966), J. Biol. Chem. 241, 2968.

Anthony, D. D., Wu, C. W; and Goldthwait, D. A. (1969),
Biochemistry 8, 246.

Bray, G. A. (1960), Anal. Biochem. I, 270.

Cardeilhac, P. T., and Cohen, S. S. (1964), Cancer Res. 24,
1595.

Chamberlin, M., and Berg, P. (1962), Proc. Nat. Acad. Sci.
U. S. 48, 81 .

Chamberlin, M., and Berg, P. (1964),]. Mol. Biol. 8, 708.

Chu, M. Y., and Fischer, G. A. (1968), Biochem. Pharmacol.
17,753.

Furth, J. J., Hurwitz, J., and Anders, M. (1962), J. Biol.
Chem. 237, 2611.

Goldberg, I. H., and Rabinowitz, M. (1961), Biochem.
Biophys. Res. Commun. 6, 394.

Holley, R. W., Apgar, J., Doctor, B. P., Farrow, J., Marini,
M. A., and Merrill, S. H. (1961), J. Biol. Chem. 236,
200.

Ikehara, M., Murao, K., Harada, E.., and Nishimura, S.
(1968), Biochim. Biophys. Acta 155, 82.

Johnson, J. C., DeBacker, M., and Boezi, J. A. (1971),
J. Biol. Chem. 246, 1222.

Johnson, J. C., Shanofi‘, M., Bass, S. T., Boezi, J. A., and
Hansen, R. G. (1968), Anal. Biochem. 26, 137.

Kahan, F. M., and Hurwitz, J. (1962), J. Biol. Chem. 237,
3778.

Krakow, J. S., and Ochoa, S. (1963), Biochem. Z. 338, 796.

Lee, L. F., and Boezi, J. A. (1966), J. Bacteriol. 92, 1821.

Lee, L. F., and Boezi, J. A. (1967),]. Viral. I, 1274.

Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall,
R. J. (1951),]. Biol. Chem. 193, 265.

Madison, J. T. (1968), Annu. Rev. Biochem. 37, 131.

Nishimura, S., Harada, F., and Ikehara, M. (1966), Biochim.
Biophys. Acta 129, 301.

Payne, K. J., and Boezi, J. A. (1970), J. Biol. Chem. 245, 1378.

Price, A. R., and Rottman, F. (1970), Biochim. Biophys. Acta
199,288.

Radding, C. M., and Kornberg, A. (1962), J. Biol. Chem.
237, 2877.

Rottman, F., and Heinlein, K. (1968), Biochemistry 7, 2634.

Rottman, F., and Johnson, K. L. (1969), Biochemistry 8,
4354.

Sentenac, A., Ruet, A., and Fromageot, P. (1968a), Eur. J.
Biochem. 5, 385.

Sentenac, A., Simon, E. J., and Fromageot, P. (1968b),
Biochim. Biophys. Acta 161, 299.

Q." nag. . .a-a‘m; 1!- ~.
-.4

”’1

 

I“

DNA POLYMERASE WITH RIBOSOMES AND MEMBRANES

Shigeura, H. T., and Boxer, G. E. (1964), Biochem. Biophys,

Res. Commun. 17, 758.
Shigeura, H. T., Boxer, G. E., Meloni, M. L., and Sampson,

S. D. '(1966), Biochemistry 5, 994.
Shigeura, H. T., and Gordon, C. N. (1965), J. Biol. Chem.

240,806.
Singh, H., and Lane, B. G. (1964), Can. J. Biochem. 42,

1011.

Slapikof‘f, S., and Berg, P. (1967), Biochemistry 6, 3654.

Thomas, C. A., and Abelson, J. (1966), in Procedures in
Nucleic Acid Research, Cantoni, S. L., and Davies, D. R.
Ed., New York, N. Y., Harper & Row, p 553.

Ts’o, P. P., Helmkamp, G. K., and Sander, C. (1962), Biochim.
Biophys. Acta 55, 584.

Weiss, S. B., and Nakamoto, T. (1961), J. Biol. Chem. 236.

618.

 

 

ART I CLE 2
TEMPLATE ACTIVITY OF 2'-0-METHYLPOLYRIBONUCLEOTIDES

WITH PSEUDOMONAS PUTIDA DNA-DEPENDENT RNA

POLYMERASE

BY

Gary F. Gerard

35

‘ 'o a. fl..‘.’o?‘l.v ~‘ ‘.

ABSTRACT

Pseudomonas pgtida RNA polymerase can use the

 

single stranded 2'-Qfmethylpolyribonucleotides, poly(Um) P
and poly(Cm), as templates for the synthesis of poly(A)

and poly(G), respectively. No template activity was

 

detected with the purine—containing 2‘-Qfmethylated homo-
polymers, poly(Am) and poly(Im). The poly(Am) strand of
either poly(Am)-poly(U) or poly(Am)-poly(Um) was not a
template for poly(U) synthesis, and did not prevent the
poly(U) or poly(Um) strand of the duplex from serving as
a template for poly(A) synthesis. The poly(Um) strand of
the duplex poly(Am)'poly(Um) was an effective template
for poly(A) synthesis, but the poly(Um) strand of poly(A)°

poly(Um) was not.

' 36

INTRODUCTION

Bacterial DNA-dependent RNA polymerase (EC 2.7.7.6)
can use single- and double-stranded polydeoxyribonucleotides r“

and polyribonucleotides as templates for the synthesis of

RNA in yitrg (l-S). A given polydeoxyribonucleotide is, I
however, a more effective template than a comparable
polyribonucleotide. Thus, RNA polymerase does not have E;
an absolute requirement for a template which has a hydrogen
at the 2'-position, but can use polynucleotides that have
a hydroxyl group at the 2'-position. Polynucleotides with
substituents at the 2'-position other than a hydrogen or
a hydroxyl group have not been tested as templates for RNA
polymerase.
As part of a general study of the structure and
_mechanism of action of Pseudomonas putida RNA polymerase
(6-8), we have undertaken a study to determine whether
or not 2'-Qfmethylhomopolyribonucleotides can serve as
templates for this enzyme. These 2'-Qfmethylated homo-
polymers have a bulkier substituent at the 2'-position
than do polydeoxyribonucleotides or polyribonucleotides.
In this report we present the results of a study of the
template activity of both single- and double-stranded

2'-Qfmethylhom0polyribonucleotides.

37

MATERIALS AND METHODS

RNA polymerase was purified from Pseudomonas putida

 

according to the procedure of Johnson, DeBacker, and Boezi

(6) with the modifications reported by Johnson (8). The T“
preparation of enzyme used in these experiments was at :
least 98% pure and contained one equivalent of sigma (o) g
per equivalent of core enzyme (dZBB'). i:
9

The synthetic polyribonucleotides used in these
studies were obtained from Miles Laboratories, Inc. The
2'-Qfmethylpolyribonucleotides were prepared according to
published procedures (9-11), with the exception of poly(Im)
(12). The synthetic polydeoxyribonucleotides were gifts
from Dr. F. J. Bollum, University of Kentucky, Lexington,
Kentucky. The mean value of the sedimentation coefficient
determined by centrifugation through sucrose density
gradients (pH 8.0) for each synthetic polyribo— and
polydeoxyribonucleotide was between 4.0 and 6.0 S. For
poly(Am), poly(Im), poly(Um), and poly(Cm), the mean sedi-
mentation coefficients were 12.4, 16.0, 10.1 and 5.1 S,
respectively. The double—stranded homOpolynucleotides
used in this study were prepared by incubating an equimolar

mixture of the complementary single-stranded polynucleotides

38

39

at room temperature in the presence of 0.01 M NaC1-0.0l M
Tris-HCl, pH 8.0, for 24 to 36 hours (13-15).

For the radioactive assay of RNA polymerase, the
standard reaction mixture (0.8 ml) for monitoring poly-

ribonucleotide synthesis contained 100 mM Tris-HCl, pH 8.0,

5 mM dithiothreitol, 15 ug/ml RNA polymerase, 50 uM of the ri
appropriate polynucleotide eXpressed in terms of nucleotide 3
phosphate, 2 mM manganese chloride, and one of the following ‘
nucleoside triphosphates: 0.4 mM 3H-ATP, 1.4 mM 3H-UTP,‘

1.2 mM 3H—GTP, or 0.4 mM 3H-CTP. The specific radioactivity %;

of the 3H-labeled nucleoside triphosphates was usually 5 to
10 x 103 cpm/nmole. The nucleoside triphosphate and the
polynucleotide template used in each type of reaction
Inixture were present at saturating concentrations. Aliquots
(50 pl) were removed from the reaction mixture at various
times during 100 minutes of incubation at 30° C. The
'trichloroacetic acid insoluble radioactivity in each
aliquot was then determined as previously described (7).
For the spectrophotometric assay of RNA polymerase
(16), the reaction mixture contained, in addition to the
cxxnponents used in the radioactive assay, 0.4 mM NADP+,
0.4 nWlUDP-glucose, and excess phosphoglucomutase, glucose-

G—Ifliosphate dehydrogenase, and UDP-glucose pyrophosphorylase.

 

RESULTS

Single-stranded homopolynucleotides as templates.
Poly(Um) was a template for poly(A) synthesis catalyzed
by E. putida RNA polymerase. The time course of the
reaction is shown in Figure la. A linear rate of poly(A)
synthesis was observed during 100 minutes of incubation.
For comparison, the time courses for the poly(dT)- and the
poly(U)-directed synthesis of poly(A) are presented. The
initial rates of AMP incorporation into trichloroacetic
acid insoluble poly(A) were 3.5, 1.5, and 0.4 nmoles/
minute per m1 of reaction mixture for poly(dT), poly(U),
and poly(Um), respectively. Thus, the initial rate of
poly(A) synthesis with poly(Um) as the template was 27%
of that observed with poly(U) as the template. No poly(Um)—
directed poly(A) synthesis was observed if 2 mM magnesium
chloride was used in place of 2 mM manganese chloride in
the standard reaction mixture.

No poly(U) synthesis as measured by the incorpora-
tion of UMP into trichloroacetic acid insoluble polymer
xvas detected with poly(Am) as the template (Figure 1b).
In experiments in which the incorporation of as little
as 0.02 nmole UMP/m1 of reaction mixture could be measured,

:10 poly(Am)-directed poly(U) synthesis was detected, even

40

41

after 100 minutes of incubation. If 2 mM magnesium chloride
was used in place of 2 mM manganese chloride in the standard
reaction mixture, again, no poly(Am)—directed poly(U)
synthesis was detected. P. putida RNA polymerase can

use poly(dA) and poly(A) as templates for poly(U) synthesis
(Figure 1b). The initial rates of UMP incorporation into
trichloroacetic acid insoluble poly(U) were 3.2 and 1.3
nmoles/minute per ml of reaction mixture for poly(dA) and
poly(A), respectively.

Even though P. putida RNA polymerase did not
catalyze the synthesis of trichloroacetic acid insoluble
poly(U) with poly(Am) as the template, the possibility
exists that the enzyme might catalyze the poly(Am)-directed
synthesis of oligonucleotides of UMP which would be soluble
in trichloroacetic acid. In order to determine if tri-
chloroacetic acid soluble oligonucleotides of UMP were
being synthesized, the spectrophotometric assay for RNA
polymerase which monitors inorganic pyrophosphate was
employed. With poly(Am) as the template and UTP as the
substrate, no inorganic pyrophosphate formation was
observed over 100 minutes of incubation. Thus, within
the limits of sensitivity of the assay (a rate of forma-
‘tion of 0.02 nmole of inorganic perphosphate/minute per
Inl of reaction mixture or greater), no poly(Am)-directed

reaction was detected, indicating that each RNA polymerase

__-..i--..n-i>.;n._-_a.n.i-L 4 _ ‘t i
‘\;

-A>~

42

molecule was not repeatedly initiating the synthesis and
release of trichloroacetic acid soluble oligonucleotides
of UMP.
P. putida RNA polymerase can bind poly(Am), even
though the enzyme can not use the 2'-gfmethylated polymer
as a template. If the standard reaction mixture contained F_
equimolar amounts of poly(Am) and poly(A), the initial 3
rate of poly(U) synthesis was 50% of that observed in the l
presence of poly(A) alone. If the poly(A)-directed
synthesis of poly(U) was allowed to proceed for 3 minutes %4
before poly(Am) was added to the reaction mixture, no 5
inhibition in the rate of poly(U) synthesis was observed.
The results obtained with the cytidine— and
inosine-containing polynucleotides are shown in Table 1.
The rate of reaction for each polynucleotide was deter-
Inined from a time course similar to those presented in
Figure 1. Poly(Cm) was a template for poly(G) synthesis.
The rate of poly(G) synthesis with poly(Cm) as the tem-
plate was slow, however, being only 3.7% of that observed
‘Nith poly(C) as the template. Within the limits of
sensitivity of the assay, no poly(Im)-directed synthesis
of poly(C) was detected. Poly(I) itself was a poor tem-
;p1ate for P. putida RNA polymerase.
Double-stranded homopolynucleotides as templates.
IDuplexes of adenosine- and uridine-containing homopoly-

:ribonuc1eotides and 2'-gfmethylhomopolyribonucleotides

43

were tested as templates for P. putida RNA polymerase. The
results are presented in Table 2. As was the case with
single-stranded poly(Am), no poly(U) synthesis was detected
with the poly(Am) strand of poly(Am)°poly(U) or poly(Am)-
poly(Um) as the template. The poly(Am) strand of the
duplex did not, however, prevent either poly(U) or poly(Um)
from serving as a template for poly(A) synthesis.

In the case of poly(Um) in a duplex, the poly(Um)
strand of poly(Am)°poly(Um) was an effective template for
poly(A) synthesis, but the poly(Um) strand of poly(A)-
poly(Um) was not. The rate of AMP incorporation with
poly(Am)-poly(Um) as the template was the same as that
observed with single—stranded poly(Um) as the template.

The rate of AMP incorporation with poly(A)°poly(Um) was
only 12% of that observed with either poly(Am)°poly(Um)

or single-stranded poly(Um) as the template.

A., w-...~——ah-—-.r LI‘ . ..Ic’.:

DISCUSSION

P. putida RNA polymerase can use the single-stranded
.2'~07methy1hom0polyribonucleotides, poly(Um) and poly(Cm), r__
as templates. Of the two pyrimidine-containing polymers,
poly(Um) was the more effective template. No template
activity was detected with the purine-containing 2'—0— g
methylated homOpolymers, poly(Am) and poly(Im). As a is
general rule, for a particular group of single-stranded
homOpolynucleotides having the same base, the order of
template effectiveness for P. putida RNA polymerase was:
polydeoxyribonucleotides > polyribonucleotides > 2'-Qf
methylpolyribonucleotides.

The finding that there was no detectable template
activity with poly(Am) but that poly(Um) and poly(Cm)

‘were templates is unexpected in light of the fact that
poly(A), like poly(U) and poly(C), is an effective tem-
plate. No poly(Im)-directed synthesis of poly(C) was
detected, but this finding is not unexpected since poly(I)
itself was a poor template for poly(C) synthesis. The
inability of poly(Am) to serve as a template for poly(U)
synthesis is probably not due to some anomaly in the

{gross secondary structure of the 2'-Qrmethylated polymer.

44

 

45

First, P. putida RNA polymerase can bind to poly(Am).
Second, there is no significant difference at 30° C and
neutral pH in the single-stranded conformation of poly(Am)
and poly(A) as measured by ultraviolet and circular
dichroism spectroscopy (13). Third, even in the duplex
structures, poly(Am)-poly(U) and poly(Am)-poly(Um), poly(Am)
does not serve as a template, but its duplex partners do.

The explanation for why poly(Am) is not a template for

“Hi. i .'

poly(U) synthesis awaits further experimentation.

-r-0|A.

The poly(Um) strand of the duplex, poly(A)-poly(Um),

4......

was not an effective template for poly(A) synthesis,
whereas the poly(Um) strand of poly(Am)-poly(Um) was.

In addition, for the duplex poly(A)-poly(Um), the poly(A)
strand was a effective template even though the poly(Um)
strand was not. These patterns of transcription are diffi-
cult to explain. Perhaps, a knowledge of the three-
dimensional structure of these duplexes will provide the
key for interpreting the observed patterns of template
activity. In this regard, x—ray diffraction studies of

the duplexes are in progress (17).

10.

ll.

12.

13.

14.

15.

REFERENCES

Chamberlin, M., and Berg, P., Proc. Nat. Acad. Sci.,
.48, 81 (1962).

 

 

Krakow, J. S., and Ochoa, 8., Proc. Nat. Acad. Sci.,
49, 88 (1963).

Straat, P. A., Ts'o, P.O.P., and Bollum, F. J., J;
Biol. Chem., 243, 5000 (1968).

 

Karstadt, M., and Krakow, J. S., J. Biol. Chem., 245,
746 (1970).

 

Fox, C. F., Robinson, W. S., Haselkorn, R., and
Weiss, S. B., J. Biol. Chem., 239, 186 (1964).

 

Johnson, J. C., DeBacker, M., and Boezi, J. A., J;
Biol. Chem., 246, 1222 (1971).

 

Gerard, G. F., Rottman, F., and Boezi, J. A.,
Biochemistry, 10, 1974 (1971).

 

Johnson, J. C., Ph. D. Thesis, Michigan State Uni-
versity, 1971.

Rottman, F., and Heinlein, K., Biochemistry, Z, 2634
(1968).

 

Rottman, F., and Johnson, K. L., Biochemistry, 8,
4354 (1969).

 

Dunlap, B. E., Friderici, K. H., and Rottman, F.,
Biochemistry, 10, 2581 (1971).

 

Rottman, F., unpublished results (1971).

Bobst, A. M., Rottman, F., and Cerutti, P. A.,
J. Mol. Biol., 46, 221 (1969).

 

Warner, R., Ann. N. Y. Acad. Sci., 69, 314 (1957).

 

Zmudzka, B., and Shugar, D., FEBS Letters, 8, 52 (1970).

 

46

l6.

17.

47

Johnson, J. C., Shanoff, M., Bass, S. T., Boezi, J. A.,
and Hansen, R. G., Anal. Biochem., 26, 137 (1968).

Langridge, R., unpublished results (1971).

48

 

F , Elli! - JV
mm.o mew AHUV>HOQ
mo.o mHU AHVNHOQ
moo.o v ceo Asvaaom
m.m mew Aupvmaom
n.m mew Auvmaom
H.o new Aauvmaom

 

HE mom mpscfle
\pmpmuomuoocfl mpmzmmozmosoe
mnemowaos: mo mmHOEc mumHUmndm mudeEmB
"wuwoon> coeuommm mwmcmmozmflme mpflmomaosz mpfluomaoscwaom

 

.mmmnmfihaom dzm mpflupm .M HOW mmpmHmEm#
mm mmpfluomaoscmaom mcflcflmucoolmcamOCH paw IwCHpflpmo UmpcmuumanmCHmnl.H mqmme

 

49

 

 

n Zilllif.flv

m.o med
m.o me: ADV>HOQ.A<V%HOQ

mo.o med
v.H me: AEDVSHOQ.A<VmHom

v.0 mefi
moo.o v can Assvsaoc.xa<cmaoc

Him meé
moo.o v can Aavsfiom.xe<vsfioc

HE mom musceE
\pmpmnomHOUCfl mumnmmOSQOCOE memHmEme

mpfimomaosc mo mmaoEc mumnumhsm mpfluomaoscmaom
”mpfloon> coflpommm mumzmmonmwue mpflmomaosz pmpcmupmlmansoo

.mmmnmfiwaom dzm

mpflusm .M new mmumHmEmu mm mmpfluomaosconflumHomoEos HmnumELml.m can mmpwu
nomHUSQOQHu>H0m0Eos mcflcwmpcoolmcflpflns Use Imcflmocmpm Umpcmuumlmansoonl.m mqmde

50

 

 

.Afidveaom paw .A<V%Hom
.A<©vmaom an Umuomnflp mflmmcucmm onxaom mo mmusoo mEHe Abe .AEDveaom
pcm .Abvmaom .Aepvwaom en pmpomufip mflmmsucem Adveaom wo mmnsoo mee Amv

II.H mmDUHm

 

51

Iw/oaivaoclaoom awn-44Q SEI‘IOWU

 

 

O O 0
<1- m (\l 9
l I I l q
’5?
L... A
< 3‘
\— >5
3‘ B
o a.
_. . . a. l
3
n V
_ 3
8
l l I ~l
l l l l
_ . —

poly(Um)

  

 

 

S
__ :r E
U o
E a.
.. 8
__6 l
s 8 8 9
Iw/oaivaociaoom dWV-HE sa1owu

 

 

4O 80

TIME ELAPSED (MINUTES)

80

4o

ARTICLE 3
RELEASE OF THE SIGMA SUBUNIT OF PSEUDOMONAS PUTIDA

DEOXYRIBONUCEIC ACID-DEPENDENT RIBONUCLEIC ACID

POLYMERASE

BY

Gary P. Gerard

52

ABSTRACT

35S-labeled Pseudomonas putida deoxyribonucleic

 

acid (DNA)-dependent ribonucleic acid (RNA) polymerase, F“
GZBB'O, was purified from cells that had been grown in ‘

a minimal medium containing sodium [358]su1fate. The

35 f

amount of S in B', B, and 0 relative to a was 3.6 to

3.6 to 2.2 to 1.0, respectively. A study of the release 3!
of the sigma subunit of P, putida RNA polymerase was

carried out following the binding of enzyme to polynucleo-

tides and during DNA-directed RNA synthesis. Sucrose

density gradient centrifugation was the technique employed

to assay for the release of 35S-labeled sigma. The subunits

35

of S-labeled RNA polymerase present in protein peaks

resolved on sucrose gradients were identified by means of

sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

35

Binding of S-labeled RNA polymerase to native DNA weakened

the interaction between sigma and core polymerase (a288')
but did not result in the release of sigma. Binding of the

35S-labeled enzyme to poly(A), poly(C), and to tRNA resulted

3SS-labeled RNA

in the release of sigma. Binding of
polymerase to poly[d(A-T)], denatured gh-l DNA, poly(dT),

and to poly(dA) did not result in the release of sigma.

53

54

Release of sigma subsequent to the binding of enzyme to
poly(dC) and poly(U) occurred in the absence of manganese
chloride but not in its presence. Sigma was released

from the enzyme-polynucleotide complex during DNA-directed
RNA synthesis. Within 3 minutes of incubation, about 60%

of the 35

S—labeled RNA polymerase molecules initiated RNA
synthesis and formed a 200 mM KCl stable complex with DNA
and nascent RNA. All or almost all of these enzyme
molecules released sigma. The other 40% of the enzyme
molecules did not form a 200 mM KCl stable complex within.
3 minutes. With longer times of incubation, these enzyme
molecules could slowly form a 200 mM KCl stable complex
but did not release sigma. The sedimentation coefficient

of E. putida sigma released during DNA—directed RNA synthe-

sis was 4.1 to 4.5 S.

INTRODUCTION

Bacterial DNA-dependent RNA polymerase (ribonucleoside.
triphosphate:RNA nucleotidyl transferase, EC 2.7.7.6) from

Escherichia coli (Burgess, 1969; Burgess et al., 1969),

 

Azotobacter vinelandii (Krakow and von der Helm, 1970),

 

and Pseudomonas putida (Johnson et al., 1971) is composed

 

of a core polymerase and a sigma subunit. The subunit
structure of the core polymerase is azBB'. The complex
between core polymerase and sigma is referred to as holo-
enzyme. Both core polymerase and holoenzyme are able to
catalyze the synthesis of RNA that is complementary to a
DNA template. With holoenzyme as the catalyst, RNA synthe-
sis i2_yitrg is initiated with high efficiency at Specific
DNA promotor sites which function in 33:9 (Bautz et al.,
1969; Sugiura et al., 1970). With core polymerase as the
catalyst, initiation of RNA synthesis does not occur
specifically at these promotor sites but occurs in a
random manner. Consequently, the RNA synthesized in yitrg
by holoenzyme corresponds more closely to RNA synthesized
[in yiyg than does that synthesized by core polymerase.

The sigma subunit functions either in the process

by which holoenzyme recognizes the specific promotor

55

“LO-«1 a... 11

v— u A- .5. 'gAsu-u
"
1

56

sites or in the process by which holoenzyme binds tightly
to them or in both processes (Hinkle and Chamberlin, 1970;
Zillig et al., 1970). Travers and Burgess (1969) have
concluded that subsequent to initiation of DNA-directed
RNA synthesis in yitrg by E. 991i RNA polymerase, sigma

is released from the enzyme-polynucleotide complex leaving
core polymerase to catalyze RNA chain elongation. The
experimental basis for this conclusion rests on the
observation that core polymerase molecules added to a
reaction mixture in which $80 DNA-directed RNA synthesis
by holoenzyme was occurring, could use sigma derived

from holoenzyme to catalyze the sigma—dependent transcrip-
tion of T4 DNA. The possibility exists, however, that
Travers and Burgess were not observing the release of
sigma from the holoenzyme—¢80 DNA—RNA complex caused by
events which are part of the RNA synthetic process. They
may have been observing the depletion of sigma from
holoenzyme—$80 DNA-RNA complex, from holoenzyme—$80 DNA
complex, or from holoenzyme itself caused by the estab-
lishment of an equilibrium exchange reaction with core
enzyme which was present at 4 times the concentration of
holoenzyme. In this regard, Travers (1971) has reported
that core enzyme molecules can rapidly exchange sigma.

In the experiments on sigma release during ¢80 DNA-directed

RNA.synthesis, Travers and Burgess (1969) did not directly

57

demonstrate the physical separation of sigma from enzyme-
polynucleotide complex nor did they measure the stoichiometry
of sigma release.

Krakow and von der Helm (1970) have presented
evidence which can be taken to mean that there is a physi-

cal separation of sigma from E. vinelandii holoenzyme

 

following the initiation of poly(A-U) synthesis from a
poly[d(A-T)] template. These workers used Fmflyacrylamide
gel electrophoresis to resolve sigma from the enzyme-
polynucleotide complex. Experiments similar to those
of Krakow and von der Helm (1970) were reported by Ruet
et a1. (1970) with E. 92E; holoenzyme and T4 DNA as the
template. The conclusions of both groups of workers can
be questioned, however, since the electrical potential
field present during polyacrylamide gel electrophoresis
could influence the interaction between sigma and core
polymerase-polynucleotide complex. In the studies of
Krakow and von der Helm (1970) and Ruet et a1. (1970),
as was the case in the study of Travers and Burgess (1969),
the stoichiometry of sigma release was not measured.
DNA—dependent RNA polymerase of E. putida is the
object of study in our laboratory. The E. putida holo-

enzyme differs from the E. coli and E. vinelandii holo-

 

enzyme in a property which may have some relevance to the
question of whether or not sigma is released. The inter-

action between sigma and core polymerase of E. coli and

1;." n A

58

E, vinelandii appears to be weaker than the interaction

 

between sigma and core polymerase from E. putida. Dis-

sociation of E. colf and E. vinelandii holoenzyme, but

 

 

not E. putida holoenzyme, to sigma and core polymerase
occurs during phosphocellulose chromatography (Burgess
et al., 1969; Johnson et al., 1971; Krakow and von der
Helm, 1970).

We have undertaken a study to determine whether
or not sigma is released from E. putida holoenzyme during
DNA-directed RNA synthesis and following the binding of
holoenzyme to a variety of polynucleotides. In this study,

358 was used. Sucrose

E, putida holoenzyme labeled with
density gradient centrifugation was employed as a technique
to resolve sigma, enzyme, and enzyme-polynucleotide complex.
Sodium dodecyl sulfate (SDS)-polyacrylamide gel electro-
phoresis was used to identify the subunits of RNA poly-
merase present in protein peaks resolved on the sucrose
gradients. The use of these techniques provides a means

of demonstrating directly the physical separation of sigma
from.core enzyme-polynucleotide complex and also provides

a means of measuring the stoichiometry of sigma release.

In this report, we present the results of this study.

MATERIALS AND METHODS

Materials. Sodium [3ssjsulfate and Omnifluor were

 

purchased from New England Nuclear. Agarose (Bio—Gel
Al.5m) was obtained from Bio-Rad Laboratories. Rifampicin
and streptolydigin were gifts from Gruppo Lepetit, Inc.,
Milan, Italy, and The Upjohn Company, respectively.
Poly(dT), poly(dC), and poly(dA) were gifts from F. J.
Bollum, University of Kentucky, Lexington, Kentucky. [3H]-
Poly(U) was from Miles Laboratories, Inc. Rabbit hemoglobin
was a gift from A. J. Morris of this department. All other
materials were obtained from sources previously described
(Johnson et al., 1971; Gerard et al., 1971).

Analytical Methods. Protein concentration was

 

determined by the method of Lowry et a1. (1951) with bovine
serum albumin as the standard. The concentration of
Pseudomonas putida bacteriophage gh-l DNA was determined

sapectrOphotometrically based on the extinction coefficient

1E§20 =200. The molar extinctions, e(P), used to determine
Exalyribonucleotide concentrations were: 10.5 x 103 at

257 nm, 9.2 x 103 at 360 nm, and 6.5 x 103 at 267 nm for

Ix31y(A), poly(U), and poly(C), respectively, in 0.1 M

IflaC1r0.05 M Tris acetate (pH 7.5) (Ts'o et al., 1962).

59

 

'l
”‘wa
n

 

60

The molar extinctions, €(P), for polydeoxyribonucleotides

were 6.7 x 103 at 260 nm and pH 7.5 for poly[d(A-T)] (Radding

and Kornberg, 1962) and 8.1 x 103, 5.3 x 103, and 9.7 x 103
at 260 nm for poly(dT), poly(dC), and poly(dA), respectively,
in 0.001 M Tris—HCl (pH 8.0) (Bollum, 1966). E. putida

RNA polymerase was assayed as previously described (Johnson
et al., 1971).

Characterization of Synthetic Polynucleotides. The
sedimentation velocity coefficients of the synthetic poly-
nucleotides used in these experiments were determined by
centrifugation through sucrose density gradients prepared
in 0.1 M KCl-0.01 M Tris-HCl (pH 8.0) with E. ggli tRNA as
the standard. Each synthetic polynucleotide had a sedi-
mentation coefficient with a mean value between 4.5 and
6.0 S. Each of the synthetic polynucleotides was an
efficient template for E. putida RNA polymerase in the
presence of manganese chloride.

Growth of Pseudomonas Putida. E. putida (the same
or similar to ATCC 12633) was grown in a medium which
contained the following in grams per liter: glucose, 20;
NH4C1, 2; Na HPO4, 6; KH2P04, 3; NaCl, 8; MgC12'6H20, 0.08;

2

iNaZSO4-10H20, 0.03, and 0.005 each of CaClz, FeC13°6H20,

Na2M004°2H20 and Mn(C2H302)2-4H20. For the production of

35S-labeled cells, 50 mCi of sodium [358]sulfate with a

specific activity of 845 mCi per mmole were added per

61

three liters of growth medium. The cells were grown at
330 on a gyrorotatory shaker in 2.8-1 Fernbach flasks

containing 500 ml of growth medium. Doubling time for
the culture was 100 minutes. The cells were harvested

at the late logarithmic phase of growth, and stored at

—200. From a three-liter culture, the yield of 358—

labeled cells was 10g (wet weight). Approximately 20 mCi

[35

of S]sulfate had been incorporated during growth of

the cells. The purification of 35S-labeled RNA polymerase

was begun one day after the 35S-labeled cells were harvested.

Purification of 35S-labeled RNA Polymerase. Frozen

 

35S-labeled E. putida (10g) and an equal amount of frozen

unlabeled E. putida cells were mixed with washed glass

beads and ground in a mortar with a pestle. After cell
rupture, RNA polymerase was purified through phosphocellulose
chromatography by the method described by Johnson et a1.
(1971) except that ASH buffer was replaced with 10 mM

Tris-HCl (pH 8.0), 10 mM MgCl 0.1 mM EDTA, 1 mM

2,
dithiothreitol, 200 mM KCl, and 15% glycerol (v/v) and
phosphocellulose chromatography was performed using
buffers which contained 15% glycerol rather than 50%
glycerol (v/v). After phosphocellulose chromatography,
Phosphocellulose Fraction I was further purified by

chromatography on an Agarose column (1.5 x 85 cm)

developed with 10 mM Tris-HCl (pH 8.0), 0.1 mM EDTA,

5. r - unzzi'm\."'.n.

.u- Ir
.

62

0.5 mM dithiothreitol, 500 mM KCl, and 5% glycerol (v/v),
followed by centrifugation through a linear 10 to 30%
glycerol gradient prepared in 50 mM potassium phOSphate
(pH 7.5), 1 mM dithiothreitol, and 200 mM KCl. The yield
of RNA polymerase was 500 ug of protein per 20g wet weight
of cells. The time elapsed between purification of a

35S-labeled enzyme preparation and the use of that

given
preparation for the eXperiments described in this report
did not exceed two weeks.

Sucrose Density Gradient Centrifugation. The

 

detailed experimental procedures used for the sucrose

density gradient centrifugation of various mixtures of

35S-labeled enzyme and polynucleotides are given in the

legends of the figures. In the fractions collected from

the sucrose gradients, the recovery of 358 was at least

70% and in most cases 80% to 90% of the 358 which had
been layered on the sucrose gradients prior to centri-
fugation. In the experiments in which mixtures of 358—
1abeled enzyme and single stranded synthetic polynucleotides
were centrifuged through sucrose gradients which contained
manganese chloride, the recovery of 358 was less than 50%
due to the fact that the 35S—labeled enzyme and 35S-labeled
enzyme—polynucleotide complexes adhered to the sides of

the centrifuge tubes. This was true for both cellulose

and polyallomer tubes (Beckman Instrument Co.). Addition

63

of bovine serum albumin to the sucrose gradients increased
.. __ 35. _ -, ,
the recovery of S to at least 90%.

SDS-Polyacrylamide Gel Electrophoresis. SDS—

 

polyacrylamide gel electrophoresis was performed using

a modification of the procedure of Shapiro et a1. (1967)
as described by Johnson et al. (1971). SDS—polyacrylamide
gels 11 cm in length were prepared from a polymerization
mixture which was 3.75% in acrylamide. For the gels that
were used in the analysis of fractions from the sucrose
gradients, the polymerization mixture contained, in addi—
tion to the ingredients previously described, 12.5%
glycerol (v/v). Following electrophoresis, the gels were
immersed in 10% trichloroacetic acid. Protein was stained
with 0.4% coomassie brilliant blue.

358 analysis was performed on both stained and
unstained gels. The gels were cut into 2 mm or 4 mm
transverse fractions using a stainless steel support

and cutting guide. Each fraction was placed in a
scintillation vial and 0.2 ml of 30% H202 was added.

After incubating at 70° for 9 hours or at 100° for 2 hours,
5 ml of a mixture containing 6 parts of Omnifluor solution
(18.1g of Omnifluor per gallon of toluene) and 7 parts of
Triton X—100 were added to each scintillation vial (Tishler
and Epstein, 1968). The vials were capped, shaken, and

35

monitored for their S content in a liquid scintillation

E...ur t-.-Au‘l—_5r~nii - :1,

-~--.

4

64

spectrometer. The recovery of 358 in the gel fractions

was in most cases at least 70% of that which had been

layered on the gel prior to electrophoresis.

RESULTS

35

Characterization of Pseudomonas Putida S-Labeled

ENA Polymerase. The specific enzymatic activity of E.

putida 35S-labeled RNA polymerase was 7800 nmoles of CMP 1

 

incorporated per hour per mg of protein at 30° using E.

putida bacteriophage gh-l DNA as the template. The specific

radioactivity of the 35S-labeled enzyme was 2.3 x 107 cpm l?

 

per mg of protein. As judged by SDS-polyacrylamide gel
electrophoresis (see Figures 1 and 2) and sucrose density
gradient centrifugation (see upper diagram of Figures 3
and 4), the 35S-labeled enzyme was at least 98% pure.

A densimetric tracing of a Coomassie brilliant
blue stained SDS-polyacrylamide gel of 35S-labeled RNA
polymerase is presented in Figure 1. As calculated from
the relative amounts of the subunits and their relative
molecular weights, there was one equivalent of sigma per
equivalent of d288' for the 35S-labeled enzyme.

358 analysis of a SDS-polyacrylamide gel of the
35S-labeled enzyme is presented in Figure 2. The amount

of 358 in 8' plus 8 and in 0 relative to a was 7.2 to 2.2

35

to 1.0, reSpectively. The amount of S of 8' was found

to be equal to that of B in three experiments in which 8'

65

66

and B were separated by SDS—polyacrylamide gel electrophoresis.
In these three experiments, the time of electrophoresis was
6 hours rather than the 4.75 hours reported for the experi-
ment presented in Figure 1.

The Release of the Sigma Subunit of RNA Polymerase
During gh-l DNA—Directed RNA Synthesis: Analysis by Centri—
fugation through Sucrose Density Gradients Containing 50 mM

KCl. Reaction mixtures containing 35S-labeled enzyme,

35S-labeled enzyme plus gh-l DNA, and 35S—labeled enzyme
plus gh-l DNA and the four common nucleoside triphosphates
were incubated for 3 minutes. After incubation, sucrose
gradient centrifugation was used to detect free 35S—labeled
sigma. If the sigma subunit is released from the holo-
enzyme during the incubation period prior to centrifugation,
it would be expected to be found near the top of the sucrose
gradient well resolved from free enzyme and enzyme-
polynucleotide complex. The results are presented in

Figure 3. 35S—labeled RNA polymerase that had been
incubated in the reaction mixture which lacked gh-l DNA

and the nucleoside triphOSphates sedimented as a single

symmetrical peak (upper diagram). 35

S-labeled RNA poly-
merase that had been incubated with gh—l DNA sedimented
much faster than free enzyme, indicating that it was bound
to gh-l DNA (middle diagram). Little or no free 358—

labeled protein was observed near the top of the sucrose

 

67

r.

the 3‘)S‘—labele<l RNA polymerase that

gradient. Most of
had been incubated with gh-l DNA and the nucleoside tri-
phosphates sedimented much faster than free enzyme (lower
diagram). This fast sedimenting material is a complex

which contained gh—l DNA, 35

S-labeled enzyme, and RNA.

RNA was determined to be present in the complex by experi—
ments monitoring radioactive RNA synthesized from 3H-
labeled nucleoside triphosphates. A small amount of 358-
labeled enzyme, which was not bound to gh-l DNA was
observed. A 35S—labeled protein peak was detected near
the tOp of the sucrose gradient (fraction numbers 24-28).
This protein peak was identified as sigma by SDS-
polyacrylamide gel electrophoresis (see below), indicating
that the sigma subunit of RNA polymerase had been released
during gh—l DNA—directed RNA synthesis. All of the RNA
polymerase molecules, however, had not released sigma.
Based on the amount of free 35S-labeled sigma and the
amount of 35S-labeled enzyme in the complex which con-
tained gh-l DNA and RNA, about 60% of the enzyme

molecules in the complex had released sigma.

35S—labeled subunits of RNA

Identification of the
polymerase present in each of the peak fractions of the
sucrose gradients was carried out by means of SDS-

polyacrylamide gel electrophoresis of comparable fractions

of duplicate sucrose gradients. In these SDS-polyacrylamide

68

gel electrophoresis experiments, 8' and 8 were not resolved
from each other. Free RNA polymerase (inset, upper diagram
of Figure 3) contained the 8'8, 0, and G subunits in the
same relative amounts as those found in enzyme that had not
been centrifuged through a sucrose gradient, i.e., one
equivalent of sigma per equivalent of QZBB'. RNA poly-
merase bound to gh—l DNA (inset, middle diagram) contained
8'8, 0, and a, but the amount of 0 relative to d and 8'8
was about 70% of that found in free enzyme. RNA poly-
merase bound in the complex that contained gh-l DNA,
enzyme, and RNA (inset A, lower diagram) contained 8'8,
0 and a. The amount of 0 was about 35% of that found in
free enzyme. The 35S-labeled protein peak near the top
of the sucrose gradient was identified as 0 (inset B, lower
diagram). Although the SDS—polyacrylamide gel data is not
shown, the enzyme not bound to gh-l DNA (lower diagram)
contained 8'8, 0 and a in the same relative amounts as
that found in enzyme that had not been centrifuged through
a sucrose gradient.

The Effect of Binding 35S-Labeled RNA Polymerase to
gh-l DNA on the Interaction between Core Polymerase and the

35

Sigma Subunit. After centrifugation, all of the S-labeled

 

RNA polymerase molecules which had sedimented as free
enzyme contained sigma (upper diagram of Figure 3), but

only 70% of the RNA polymerase molecules which had

69

sedimented as enzyme-gh-l DNA complex contained the subunit
(middle diagram of Figure 3). Sigma which had dissociated
from 30% of the RNA polymerase molecules of the enzyme—gh-l
DNA complex was not detected as a peak near the top of the
sucrose gradient. This suggests that dissociation of sigma
occurred during centrifugation as the enzyme-gh-l DNA
complex moved through the gradient. Accordingly, sigma
should be found trailing behind the enzyme—gh-l DNA complex
throughout the sucrose gradient.

The concentration of KCl in the sucrose gradients
affected the dissociation of sigma from the enzyme-gh—l
DNA complex. When 0, 50, or 100 mM KCl was present in the
gradients, the enzyme-gh-l DNA complex was stable, but
sigma dissociated during centrifugation from 60, 30, and
10%, respectively, of the enzyme molecules of the enzyme-
gh-l DNA complex. When 200 mM KCl was present in the
gradient, the enzyme-gh-l DNA complex was not stable and
diSsociation of the complex occurred (see middle diagram
of Figure 4). In the experiment in which no KCl was
present in the sucrose gradients, SDS polyacrylamide gel
analysis showed that sigma trailed behind the enzyme-gh-l
DNA complex throughout the gradient. When 0, 50, 100, or
200 mM KCl was present in the gradients, sigma did not
dissociate during centrifugation from RNA polymerase

molecules which had not been reacted with gh-l DNA.

. . w“"-.‘fi

,m._ L. ”I

 

70

The Release of the Sigma Subunit of RNA Polymerase

 

during gh-l DNA-Directed RNA Synthesis: Analysis by Centri-

fugation through Sucrose Density Gradients Containing 200
35

 

mM KCl. Based on the amount of free S-labeled sigma
found near the top of the sucrose gradient which contained
50 mM KCl (lower diagram of Figure 3) and the amount of
sigma found in the complex that contained gh-l DNA, enzyme,
and nascent RNA (inset A, lower diagram of Figure 3), only
about 60% of the enzyme molecules in the complex released
sigma. The other 40% of the enzyme molecules in the complex
did not release sigma. Presumably, the enzyme molecules
which released sigma were engaged in RNA synthesis and

those which did not release sigma were not. To distinguish
between enzyme molecules engaged in RNA synthesis and those
not engaged in RNA synthesis, centrifugation through sucrose
gradients containing 200 mM KCl was used. The ternary
complex between RNA polymerase, DNA and nascent RNA is
stable in solutions of high ionic strength, but the binary
complex between RNA polymerase and DNA is not (Richardson
1966a; l966b).

35

Reaction mixtures containing S—labeled enzyme,

35 35

S-labeled enzyme plus gh-l DNA, and S—labeled enzyme
plus gh—l DNA and the four common nucleoside triphosphates
were incubated for 3 minutes and then analyzed by centri-

fugation through sucrose density gradients containing

 

71

200 mM KCl. The results are presented in Figure 4. 358-

labeled RNA polymerase that had been incubated in the
reaction mixture which lacked gh-l DNA and the nucleoside
triphosphates sedimented as a symmetrical peak with a
sedimentation coefficient of 13 S (Johnson et al., 1971)
(upper diagram). 35S-labeled RNA polymerase that had
been incubated with gh-l DNA to promote binding of the
enzyme, indicating that the enzyme dissociated from gh-l
DNA as it moved into the sucrose gradient containing 200 mM
KCl (middle diagram). About 40% of the 35S-labeled RNA
polymerase molecules that had been incubated with gh-l
DNA and the nucleoside triphosphates sedimented as free
enzyme at 13 S and the remaining 60% sedimented in a 200
mM KCl stable complex that sedimented faster than 13 S.
Free 35S-labeled sigma whose identification is described
below was observed near the top of the gradient.
Identification of the 35S-labeled subunits of RNA
polymerase present in each of the peak fractions was carried
out using SDS—polyacrylamide gel electrophoresis. Both
free enzyme and enzyme that had been bound to gh—l DNA
but dissociated from it in the sucrose gradient contained
a full compliment of sigma, i.e., one equivalent of sigma
per equivalent of d288' (inset, upper and middle diagram
of Figure 4). 35S-labeled enzyme found in the 200 mM KCl

stable complex contained only about 10% of the sigma found

72

in free enzyme (inset A, lower diagram). Free 35S—labeled

enzyme at the 13 S position of the sucrose gradient con-
tained a full compliment of sigma (inset B, lower diagram).
The 35S-labeled protein near the top of the gradient was
identified as sigma (inset C, lower diagram). The amount

of the free 35

S-labeled sigma near the top of the gradient
was equivalent to the amount of sigma released from the
enzyme in the 200 mM KCl stable complex.

The 200 mM KCl stable complex contained 35S-labeled
RNA polymerase molecules that were engaged in gh-l DNA-
directed RNA synthesis. Radioactive RNA synthesized from
3H-labeled nucleoside triphosphates cosedimented with the

35S-labeled RNA polymerase mole-

complex but not with the
.cules that sedimented as free enzyme. No 200 mM KCl stable
complex was formed if either gh-l DNA or the four common
nucleoside triphosphates were omitted from the reaction
mixture. If the reaction mixture contained ATP rather

than all four of the nucleoside triphosphates, only about
8% of the enzyme molecules formed a 200 mM KCl stable
complex (upper left diagram of Figure 5). If rifampicin
was added to the complete reaction mixture, about 8% of the
enzyme molecules formed a 200 mM KCl stable complex (upper
right diagram of Figure 5).

The effects of the addition of streptolydigin and

RNase to the complete reaction mixture on the formation of

 

73

the 200 mM KCl stable complex were examined. Streptolydigin
is an antibiotic which inhibits RNA synthesis by RNA poly-
merase by inhibiting the rate of phosphodiester bond forma-
tion (Cassani et al., 1971). If streptolydigin was added
to a complete reaction mixture, about 18% of the enzyme
molecules formed a 200 mM KCl stable complex after 3
minutes of incubation (lower left diagram of Figure 5).

If RNase was added to the complete reaction mixture, about
50% of the enzyme molecules formed a 200 mM KCl stable
complex and released sigma (lower right diagram of Figure
5). This result is similar to that obtained in the absence
of RNase (see lower diagram of Figure 4). Thus, hydrolysis
by RNase of the nascent RNA chain protruding from the
enzyme surface did not effect the formation of the 200 mM
KCl stable complex or the release of sigma.

The effects of varying the concentration of RNA
polymerase and the time of incubation on the amount of
enzyme in the 200 mM KCl stable complex and the amount
of free sigma found near the top of the sucrose gradient

353—1abe1ed RNA

were examined. When 8, 25, and 50 ug of
polymerase per m1 of reaction mixture was used, the amount
of enzyme in the 200 mM KCl stable complex and the amount
of free sigma increased proportionately. In this concen—

tration range, the rate of gh-l DNA-directed RNA synthesis

is proportional to enzyme concentration. When the complete

74

reaction mixture was incubated 10 and 20 minutes rather
than the standard 3 minute incubation, the amount of
enzyme in the 200 mM KCl stable complex increased from
60% of the total enzyme to 70% and 80%, respectively.

No increase in the amount of free sigma found near the
top of the sucrose gradient was detected with the longer
times of incubation.

Determination of the Sedimentation Coefficient of

 

the Sigma Subunit of RNA Polymerase Released during gh-l

 

DNA-Directed RNA Synthesis. The sedimentation coefficient

 

of sigma was determined using centrifugation through sucrose

density gradients containing 50 mM KCl. Escherichia coli

 

alkaline phosphatase with a sedimentation coefficient of

6.3 S (Garen and Levinthal, 1960) and rabbit hemoglobin
with a sedimentation coefficient of 4.2 S (Chiancone et al.,
1966) were used as markers. The results are presented in
Figure 6. Free sigma sedimented as a nearly symmetrical
peak with a sedimentation coefficient of 4.1 to 4.5 S.

35

The Effect of Binding S-Labeled RNA Polymerase

 

Eg_Polyribonucleotides on the Release of the Sigma Subunit.
35

 

The effect of binding S-labeled RNA polymerase to several
polyribonucleotides on the release of the sigma subunit was
examined. The results with poly(U) are presented in Figure
7. 35S-labeled RNA polymerase was incubated with [3H]poly(U)

in the presence of manganese chloride to promote the binding

75

of the enzyme to the polyribonucleotide, and was then
centrifuged through a sucrose gradient which contained
manganese chloride. No peak of sigma was detected near
the top of the gradient (upper diagram of Figure 7),
indicating that the binding of enzyme to poly(U) in the
presence of manganese chloride did not result in the
release of the sigma subunit. The binding of the enzyme 3
to poly(U) in the absence of manganese chloride, however,
resulted in the release of sigma. If binding and centri-

fugation were carried out in the absence of manganese

 

chloride, free sigma was detected near the top of the
sucrose gradient (lower diagram of Figure 7). About 70%
of the enzyme molecules released sigma.

The binding of enzyme to poly(C), to poly(A),
and to tRNA in either the presence or absence of manganese
chloride resulted in the release of the sigma subunit. In
these experiments, 40 to 60% of the enzyme molecules
released sigma. In Figure 8, the results are presented
for the binding of the enzyme to poly(C) in the presence
and in the absence of manganese chloride (upper and middle
diagram) and for poly(A) in the presence of manganese
chloride (lower diagram). The data for tRNA are not
presented. In the case of poly(A), SDS-polyacrylamide
gel electrophoresis analysis showed that about 60% of

the enzyme molecules in the main sedimentation peak and

76

in its fast moving shoulder had released sigma and that
the 35S—labeled protein that was detected near the tOp
of the sucrose gradient was sigma (insets, lower diagram

of Figure 8).

The Effect of Binding 35S-Labeled RNA Polymerase

 

to Polydeoxyribonucleotides on the Release of the Sigma

 

Subunit. As already described, the binding of 35S-labeled
RNA polymerase to native gh-l DNA did not result in the

35

release of the sigma subunit, i.e., no peak of S-labeled

sigma was observed near the top of the sucrose gradient
(middle diagram of Figure 3). The effect of binding 358-
labeled RNA polymerase to other polydeoxyribonucleotides
on the release of the sigma subunit was examined. The
results are presented in Figure 9 for denatured gh-l DNA
(left diagram) and for poly[d(A-T)] (right diagram). In
these sucrose density gradients, no peak of free sigma

35S-labeled

was observed, indicating that the binding of
RNA polymerase to these polydeoxyribonucleotides did not
result in the release of the sigma subunit. In the case
of denatured gh-l DNA, SDS-polyacrylamide gel electro-
phoresis analysis showed that RNA polymerase complexed

to denatured gh-l DNA and centrifuged through the sucrose

gradient contained a full compliment of sigma (inset,

left diagram of Figure 9).

77

In Figure 10 the results of the binding of 358-

labeled RNA polymerase to poly(dT) (left diagram) and to
poly(dC) (right diagram) are presented. In these experi-
ments, manganese chloride was present in both the reaction
mixtures and in the sucrose gradients. No release of
sigma was observed. For poly(dC) but not for poly(dT),
release of sigma was observed in the absence of manganese
chloride (data not shown). The binding of the enzyme to
poly(dA) in the presence or absence of manganese chloride
did not result in the release of sigma (data not shown).
In the studies of the release of the sigma subunit
during gh-l DNA-directed RNA synthesis, the reaction mix—
ture contained divalent metal ions, but the 50 mM and 200
mM KCl sucrose gradients did not (see Figures 3 and 4).
The inclusion of manganese or magnesium chloride in the
sucrose gradients, however, did not alter the amount of

free sigma found near the top of the sucrose gradients.

 

DISCUSSION

35S—labeled Pseudomonas putida RNA polymerase,

 

o288'o, of high specific radioactivity was purified from
35S—labeled cells that had been grown on a minimal growth
medium which contained sodium [3SSjsulfate. As a result
of the high specific radioactivity of the enzyme, the
assay of as little as 0.01 pg of protein was feasible by
monitoring 358.

Analysis of the E. putida RNA polymerase subunits
showed that the 358 content of 8', 8, and 0 relative to a
was 3.6 to 3.6 to 2.2 to 1.0. Since the molar concentra-
tions of 8', B and 0 relative to o are 0.5 to 0.5 to 0.5
to 1.0 (Johnson et al., 1971), there are 7.2 sulfur atoms
in 8' and in 8 and 4.4 sulfur atoms in O for each sulfur
atom in d. Consequently, in each 44,000g of B', of 8, and
of 0, there are twice as many sulfur-containing amino acids

as in a gram molecular weight of a which is 44,0009.

Burgess (1969) reported that for Escherichia coli RNA

 

polymerase, in each 39,000g of 8' plus 8 there are 1.3
times as many sulfur-containing amino acids as in a gram
molecular weight of a which is 39,000g. The values for

E. coli 0 and for 8' and 8 separately have not been reported.

78

 

79

Sigma was released from the enzyme—polynucleotide
complex during gh-l DNA—directed RNA synthesis. Within 3
minutes of incubation, about 60% of the 35S-labeled RNA
polymerase molecules initiated RNA synthesis and formed a
200 mM KCl stable complex with the gh-l DNA template and
the growing RNA chain. All or almost all of these enzyme
molecules released sigma. The other 40% of the enzyme
molecules did not form a 200 mM KCl stable complex within
3 minutes of incubation because, presumably, they did not
initiate RNA synthesis. With longer times of incubation,
these enzyme molecules could slowly form a 200 mM KCl stable
complex but did not release sigma. These enzyme molecules
are considered to be defective in their capacity to synthe-
size RNA because they initiate RNA synthesis very slowly.
The enzyme molecules that finally do initiate RNA synthesis
and form a 200 mM KCl stable complex with the longer times
of incubation do not release sigma either because they have
a direct impairment in the sigma release process itself,
or because some event in RNA synthesis which triggers
sigma release does not occur. Perhaps, sigma release is
triggered once the synthesis of nascent RNA of a certain
nucleotide length is achieved (Travers, 1971). If nascent
RNA of this nucleotide length was not synthesized due to
the fact that the defective enzyme molecules were impaired
with respect to their ability to carry out phosphodiester

bond formation, sigma would not be released.

80

The defective RNA polymerase molecules probably

were not produced by the decay of 358. At the level of

35S-labeling of RNA polymerase, only one enzyme molecule
in every 50 to 100 contained a 35S atom. Enzymatic activity

35

of S-labeled RNA polymerase was stable for 90 days when

stored at -20° which is slightly longer than the 87.9 day E“)
half-life of 358. The defective enzyme molecules may i l
have been damaged during the enzyme purification procedure. )

35S-labeled

Although the specific enzymatic activity of the
RNA polymerase preparation was comparable to most prepara- p’
tions of unlabeled E. putida RNA polymerase, enzyme prepara-
tions of higher Specific enzymatic activity have been
obtained by us.
The sedimentation coefficient of E. putida sigma
released during gh—l DNA—directed RNA synthesis was 4.1
to 4.5 S. The molecular weight of E. putida sigma is
98,000 (Johnson et al., 1971). Enzymes of globular confor-
mation with molecular weights in the 100,000 range usually
have sedimentation coefficients of 5.5 to 6.0 S (Holleman,
1966). Examples of enzymes involved in nucleic acid
metabolism which are composed of single polypeptide chains
of molecular weights in the 100,000 range and which have
sedimentation coefficients of 5.5 to 6.0 8 have been

reported by Baldwin and Berg (1966), Jovin et al. (1969),

Yaniv and Gros (1969), and Hayashi et al. (1970). The

81

relatively low sedimentation coefficient for E. putida
sigma indicates that it does not have a globular confor-
mation. The conformation of E. 22$; sigma probably is
also not globular. Its molecular weight and sedimentation
coefficient are 95,000 and 4.5 to 5.0 S, respectively

(Burgess et al., 1969).

In the experiments in which the binding of E. putida
35S-labeled RNA polymerase to polynucleotides was examined,
the results obtained in the presence of manganese chloride
should be emphasized since RNA synthesis is possible only
in the presence of a divalent metal ion. Under this condi-
tion, binding of the enzyme to polydeoxyribonucleotides did
not result in the release of sigma. Binding of the enzyme
to polyribonucleotides on the other hand, with one
exception, did result in the release of sigma.

Krakow and von der helm (1970) have reported that
for Azotobacter vinelandii RNA polymerase, binding of the
holoenzyme to single-stranded polydeoxyribonucleotides
and polyribonucleotides caused the release of sigma. In
their studies, Krakow and von der Helm used electrophoresis
in polyacrylamide gels to assay for the release of sigma
from the enzyme-polynucleotide complex. The effects that
this method might have on the interaction between sigma

and core-polynucleotide complex cannot be defined or

controlled. For example, the effects of monovalent and

 

___

A

Jaw-n- axe-y -'.M‘-
I"

82

divalent cations which we have described in this study
could not be appreciated using the polyacrylamide gel
electrophoresis technique.

The mechanism by which sigma is released from the
enzyme—polynucleotide complex following the binding of
enzyme to polyribonucleotides is not understood. We do
not know where on the enzyme surface the polyribonucleo-
tides bind to affect the release of sigma. They could
bind to the enzyme at the template binding site usually
occupied by DNA or at the nascent RNA binding site
occupied by the growing RNA chain during DNA-directed
RNA synthesis. In the eXperiments in which the binding
of E. putida RNA polymerase to synthetic polyribonucleo-
tides was shown to result in the release of sigma, in
fact, usually only about 60% of the enzyme molecules
released sigma. Perhaps, the enzyme molecules which
did not release sigma on binding to the synthetic poly-
ribonucleotides were the defective enzyme molecules
which could not release sigma during gh-l DNA-directed
RNA synthesis. Further experimentation is required
before the mechanism of sigma release following the
binding of the enzyme to polynucleotides and during

DNA—directed RNA synthesis is understood.

REFERENCES

Baldwin, A., and Berg, P., J. Biol. Chem.; 241, 831 (1966).

 

Bautz, E. K. F., Bautz, F. A., and Dunn, J. J., Nature,
223, 1022 (1969).

Bollum, F. J., in Procedures in Nucleic Acid Research,
S. L. Cantoni and D. R. Davies (Editors), Harper
and Row, New York, 1966, p. 579.

 

Bray, G. A., Anal. Biochem., l, 270 (1960).

 

Burgess, R. R., J. Biol. Chem., 244, 6168 (1969).

 

Burgess, R. R., Travers, A. A., Dunn, J. J., and Bautz,
E. K. F., Nature, 221, 43 (1969).

 

Cassani, G., Burgess, R. R., Goodman, H. M., and Gold, L.,
Nature New Biology, 230, 197 (1971).

 

Chiancone, E., Vecchini, P., Forlani, L., Antonini, E.,
and Wyman, J., Biochim. Biophys. Acta, 127, 549
(1966).

Garen, A., and Levinthal, C., Biochim. Biophys. Acta, EE,
470 (1960).

 

Gerard, G. F., Rottman, F., and Boezi, J. A., Biochemistry,
lg, 1974 (1971).

 

Hayashi, H., Knowles, J. R., Katze, J. R., Lapointe, J.,
and 8011, D., J. Biol. Chem., 245, 1401 (1970).

 

Hinkle, D. C., and Chamberlin, M., Cold Spring Harbor Symp.
Quant. Biol., XXXV, 65 (1970Y.

 

Holleman, W. H., Ph.D. Thesis, Michigan State University,
1966.

Johnson, J. C., DeBacker, M., and Boezi, J. A., J. Biol.
Chem., 246, 1222 (1971).

 

83

84

Jovin, T. M., Englund, P. T., and Bertsch, L. L., J. Biol.
Chem., 244, 2996 (1969).

Krakow, J. S., and von der Helm, K., Cold Spring Harbor
Eymp. Quant. Biol., XXXV, 73 (1970).

 

Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall,
R. J., J. Biol. Chem., 193, 265 (1951).

Radding, C. M., and Kornberg, A., J. Biol. Chem., 237,
2877 (1962).

 

 

 

 

Richardson, J. P., J. Mol. Biol., El, 83 (1966a). f
Richardson, J. P., J. Mol. Biol., EE, 115 (1966b). ;
Ruet, A., Sentenac, A., and Fromageot, P., FEBS Letters, E

El, 169 (1970). 3
Shapiro, A. L., Vinuela, E., and Maizel, J. V., Biochem. ;

Biophys. Res. Commun., EE, 815 (1967).

 

Sugiura, M., Okamoto, T., and Takanami, M., Nature, 225,
598 (1970).

 

 

Tishler, P. V., and Epstein, C. J., Anal. Biochem., EE,
89 (1968).

Travers, A., Nature New Biology, 229, 69 (1971).

_

 

Travers, A. A., and Burgess, R. R., Nature, 222, 537 (1969).

 

Ts'o, P. P., Helmkamp, G. K., and Sander, C., Biochim.
Biophys. Acta, E2, 584 (1962).

 

Yaniv, M., and Gros, F., J. Mol. Biol., 33, 1 (1969).

 

Zillig, W., Zeckel, K., Rabussay, D., Schachner, M.,
Sethi, V. 8., Palm, P., Heil, A., and Seifert, W.,
923d Spring Harbor Symp. Quant. Biol., XXXV, 47
(1970).

85

 

 

 

.c ccc c .m ..m acme“ cu uemfi

Eonw mum muecsndm one .wnommcmuu Hmmcfla UHOMHAO m mcflms Es omm
no mcmE mm3 mcHUMHu oeumeflmcmp esp can mafia ucmHHHflHQ mflmmmEoou
Suez pmcflmum mmz cflmuoum one .Hmm\mE m pm meson me.v How omm

um meHOwumm mmB mflmmuosmonuomam .Hmm mpHEmamuomeaomlmom EU HH
6 co pmnmwma was Acflmuonm Mo on NV musprE may mo mHQEmm Halv m

.A>\>V Houmcsam em ccm .Hocmcumoumccucsum wo.a .mom mo.H .AH.e mac

mumzmmoge ESAUOm z H.o mCHchpcoo COHDSHOm mo H: om Ce mopsaHE OH
new coca um pmumnsocfl mmz Am: omv mmmumEmaom mzm pmeQmHlmmm
.mmmsmezaom «2m pmawnmqlmmm
mo How mpflEmH>H0m>Homlmom pmcemum
mSHm ucmeaaflum mflmmmEooo m mo mcflomue 0flHumEHmcmoll.H mmDon

86

 

 

 

 

 

 

 

“I“ 099 IV BONVBHOSBV

DISTANCE FROM ORIGIN

 

l.|illl|llllll.l!lllllill

87

[I'cl'llnll‘ll‘ll‘l’ll
'Il'

.mponuoz one mHoHHoumz CH
ponflnomop ponuoE onu mcfims ponwaocm mmz noflpooum nomo wo
unounoo mmm one .nomo ES N mo onceuoonm omuo>mnouu once

poo mm3 H ousmem CH noncomoum mneomuu oeuuofiflmnoc ony
mn pouwamcm mma nonnz ommuoEmHom nzm poaonoalmmm mo Hom
opHanwnoomaomlmom ponflmum oDHn unmeaaflnn oflmmoEoou one

.omonoemaom nzm poaonoqummm

me How opefioawnoowaomnmom o mo memeaonm mmm||.m mmDUHm

 

88

mum—232 zo_._.o<mn_

 

 

 

 

 

 

 

 

 

N.

(ho: x mm) s“

89

FIGURE 3.--The Release of the Sigma Subunit of RNA Polymerase

during gh-l DNA Directed RNA Synthesis: Analysis by Centri-

fugation through Sucrose Density Gradients Containing 50 mM
KCl and by SDS—Polyacrylamide Gel Electrophoresis.

(0.12 ml) which contained 20 mM Tris-HCl
(pH8.0), 4 mM MgC13, 1 mM MnClz, 1 mM dithiothreitol, 20 mM _
KCl, and 25 ug/ml SS-labeled RNA polymerase (upper diagram),
or the above constituents plus 100 ug/ml gh-l DNA (middle
diagram) or plus 100 ug/ml gh-l DNA and 0.4 mM each of ATP,
GTP, CTP, and UTP (lower diagram) were incubated at 300 for
3 minutes. After incubation, the reaction mixtures were
cooled to 4°. A sample (0.10 ml) from each reaction mix-
ture was then layered on a sucrose density gradient pre-
pared by layering a 4.2 ml 5 to 20% linear sucrose gradient
on top of 0.8 ml of 50% sucrose. The sucrose was prepared
in 50 mM Tris-HCl (pH 8.0), 50 mM KCl, and 1 mM dithio-
threitol. After centrifugation at 4° for 7 hr at 35,000
rpm in the Spinco SW39 rotor, fractions (0.16 ml) were
collected from the bottom of the centrifuge tubes. Each
fraction was then assayed for 35S by liquid scintillation
spectrometry (Bray, 1960). Comparable fractions from
duplicate sucrose density gradients were made 200 mM in
KCl and were incubated at 100° for 15 min in 200 pl of a
solution containing 0.1 M sodium phosphate (pH 7.1), 1.0%
SDS, and 1.0% 2-mercaptoethanol. The total SDS-treated
mixtures were layered on 11 cm SDS—polyacrylamide gels.
Electrophoresis was then performed at 25° at 4 ma/gel for
the first 15 min, and then at 8 ma/gel for the remainder
of the time. Total time of electrophoresis was 4.25 hr
for sucrose density gradient fraction number 18 (inset,
upper diagram) and for fraction number 27 (inset B, lower
diagram) and was 5.25 hr for fraction number 6 (inset,
middle diagram) and fraction number 4 (inset A, lower
diagram). After electrophoresis, the gels were cut into

4 mm transverse fractions and the 358 content of each gel
fraction was analyzed according to the procedure described
in Materials and Methods. For each inset, the ordinate

is expressed as 358-CPM x 10"2 and the abscissa is
expressed as Gel Fraction Number. The concentration of
sucrose in the SDS-treated mixtures which were layered on
the gels affected the electrophoretic mobilities of the
358-labeled subunits of RNA polymerase. Identification

of the RNA polymerase subunits in the sucrose gradient
fractions was made by comparison to the subunits of

enzyme that had not been centrifuged through a sucrose
gradient, but otherwise subjected to SDS-polyacrylamide
electrophoresis in an identical manner.

Reaction mixtures

9O

 

 

 

 

 

 

 

 

 

 

 

_ _

 

 

 

 

 

 

 

 

 

8 4
to. x 2%.?»

FRACTION NUMBER

91

FIGURE 4.--The Release of the Sigma Subunit of RNA Polymerase

during gh-l DNA-Directed RNA Synthesis: Analysis by Centri-

fugation through Sucrose Density Gradients Containing 200 mM
KCl and by SDS-Polyacrylamide Gel Electrophoresis.

Reaction mixtures identical to those described for the upper,
middle, and lower diagrams in the legend to Figure 3 were
incubated at 30° for 3 min. After cooling, a sample (0.10
ml) from each reaction mixture was layered on a sucrose
density gradient prepared by layering a 4.2 ml 5 to 20%
linear sucrose gradient on top of 0.8 ml of 50% sucrose.
The sucrose was prepared in 50 mM Tris-HCl (pH 8.0), 200
mM KCl, and 1 mM dithiothreitol. After centrifugation at
4° for 7 hr at 35,000 rpm, fractions were collected from
the bottom of the centrifuge tubes and assayed for 35S

by liquid scintillation spectrometry (Bray, 1960).

Peak fractions from duplicate sucrose gradients were
denatured in SDS as described in the legend to Figure 3

and then layered on 11 cm SDS-polyacrylamide gels.
Electrophoresis was at 25° at 4 ma/gel for the first 15
min, and then at 8 ma/gel for the remainder of the time.
Total time of electrOphoresis was 4.25 hr for sucrose
gradient fraction 20 (inset, upper diagram), fraction
number 18 (inset, middle diagram), and fraction number

18 (inset B, lower diagram), and was 5.25 hr for fraction
number 6 (inset A, lower diagram) and fraction number 27
(inset C, lower diagram). After electrophoresis, the gels
were cut into 4 mm transverse fractions and the 358 content
of each gel fraction was analyzed according to the procedure
described in Materials and Methods. For each inset, the
ordinate is expressed as 35S-CPM x 10"2 and the abscissa

as Gel Fraction Number.

 

 

 

 

 

92

 

 

 

 

 

 

 

 

 

 

 

3'5s-cm x I ’3

 

 

 

 

 

 

 

 

 

 

 

 

 

l l I
0 IO 20 30
FRACTION NUMBER

 

 

 

93

iili|iI||IFIEIll11l|llllllillfi

.oHDHxHE noHuoooH onu
nH nHmemHOHmonvm SE mm.o no CHOHQEMHHH HE\m: H HonuHo mo
oonomoum onu nH oumu Houunoo onp Ho>o wmm an commouoop mm3
dzm ownH mzuhmmg Ho COHuonomuoonH one an ponsmmoE mm dzm Ho
memccecsm cmuomuflcuazc Hucm one do once one .4 ounces on
pnomoH onu nH ponHHomop mm HUM 2E oom mcHnHMHnoo mpnoHcmnm
omOHUSm no poumHmnm mm3 ousuxHE EOHuomou nooo mo AHE OH.oV
onEom m .EOHumndonH Houmd .nHE m mom com um ponmnsonH

onoB nEmHmMHp uanu uoonV ommzm UHuoouonom HE\m: m msHm Ho
.AEoumoHp umoH Hoonv nHmeeHonmonum 2E mm.o mDHm .AEonoHp
Hanu Hommsv nHonEoHHH HE\ma H.msHm me: one .meo .mew

.med mo nomo SE «.0 no .nEonMHp umoH Hommsv med 2E w.o one
.dzo Hlnm HE\m1 OOH .ommnoEmHom dzm poHonoHlmmm HE\mn mm .HUM
:5 om .HouHounuoHnuHc as H .mHocz :5 H .NHomz as e .no.m may
HomlmHHe 2E om ponHmpnoo noHn3 AHE NH.ov mousuxHE COHuooom

.onmEou oHnoum HUM 2E com o mo COHHoEHom onu no omozm
one .nHmemHoumoHHm .cHonEmem .med mo poomwm onell.m mmDon

94

.u‘
o . vo.~'.-nl‘l.lllr. . nil in VI

mwmiaz 29.841...

 

 

 

 

 

 

 

 

Om ON 0.
H H
omozm 2.0.940Emmem
P _ e —
H
II
Z_o_as_<u_m a._.<
— _ _ _ L

 

 

8

IO
N

95

.HIV En own no oonmnHOmnm
mnHHSmmoE Sn conHEHopop mm3 anonoEon annmu Ho GOHpHmom

one . Adv ouonmmonm HSnonmouannm ouounmnnm onH mnHm: En

OHH no SHHmoHHuoEouonmouuoomm pommmmm mos ommpmnmmonm onHHome
HHoo .m .nome .Smumv SnuoEonuoomm nOHumHHHHnHUm pHsoHH Sn
AOV mmm How pohmmmw moz omonoEeHom dzm poHonMHlmmm manHmu
unoo unoHcmnm onp Eonw GOHHUMHH nomm .monSH omdeHHnoo onn

mo Eouuon onu Eoum pouooHHoo ouo3 SHE wH.ov mQOHuomuw .Houou
HomZm occHem c cH any ooe.em um us a “on 0e um coHucmcoHuucmo
Houmd .nHEdnHm Eduom onH>on HE\mn ooH one .HouHoHnuOHnqu SE H
.HUS SE om .Ao.m mac HumlmHHe SE om nH pouomonm mo3 omOHOSm one
.omonosm wme mo HE m.o mo mo» no poHoSoH noon con noHn3 omonosm
wom mo HE v.0 mo mou no unoHpoum omOHUSm HoonHH wow on m HE m.m
o manoSoH Sn couomonm unoHpmum omOHUSm o no poHoSoH mo3 ounprE
n0HuomoH nooo Eoum AHE 0H.ov oHdEmm d .poHHHno nonp cam com um
mouscHE m mom coumnsonH oHoB anonoEon pHnnmu HE\mans ovmd mm
no .ommuonmmonm onHHoxHo HHoo .m HE\mn one no .omoHoESHom dzm
cmHmncH-mmm Hs\ma OH ccc .dzo H-nm Hs\ma OOH .meo ccc .meo .meo
.med mo nomo SE «.0 HonuHo one .nHEano Ednom onH>on HE\m1 omm
.Hor SE om .HonHonnuoHnuHc SE H .mHocS SE H .mHomS :2 H .Ao.m
rev Homanue SE om conHmunoo noHn3 AHE NH.ov monsuxHE COHuomom

 

 

.mHmmcuc>m dzm conumuHaudzo anm
mnHunp commoHom omouoESHom dzm Ho pHnsndm oEmHm
onp mo pnoHOHHHoOU COHpounoEHpom onp Ho QOHHonHEHoquII.m mmDUHm

96

w“ 0179 IV BONVGHOSBV

IQ V: "2 N. -.
I I I I I I j
W“ Ol‘b .LV BONVQHOSBVV

IQ <1: IQ N -.
l I I I I

 

 

 

 

 

 

 

30

I5 20 25
FRACTION NUMBER

IO

5

97

FIGURE 7.—-The Effect of Binding 35S-Labeled RNA Polymerase
to [3H]Poly(U) on the Release of the Sigma Subunit or RNA
Polymerase.

Reaction mixtures (0.12 ml) which contained 100 mM Tris-HCl
(pH 8.0), 1 mM dithiothreitol, 100 um [3H]poly(U) (5.8 x
105 Cpm/umole), 250 ug/ml bovine serum albumin, 10 ug/ml
35S-labe1ed RNA polymerase, and either 2 mM MnClz (upper
diagram) or no added MnClz (lower diagram) were incubated
at 30° for 3 minutes and cooled. A sample (0.10 ml) from
each reaction mixture was layered on a sucrose gradient
prepared by layering a 4.2 ml 5 to 20% linear sucrose
gradient on top of 0.8 ml of 50% sucrose. The sucrose

was prepared in a buffer containing 50 mM Tris-HCl (pH
8.0), 50 mM KCl, 1 mM dithiothreitol, 100 ug/ml bovine serum
albumin, and either 2 mM MnClz (upper diagram) or no added
MnC12 (lower diagram). After centrifugation at 4° for 5.2
hr at 50,000 rpm in the Spinco SW50L rotor, fractions (0.16
ml) were collected from the bottom of the centrifuge tubes.
Each fraction was then assayed for 358 (Q) and 3H (A or
I) by liquid scintillation spectrometry (Bray, 1960).

98

HIV 0. X Soul...»

 

 

 

 

 

 

 

 

 

 

”I
nlo P. 4 B m/I. 8 4
_ _ _ _ _ H _ n _
8 6 2 8 6 4 2
_ _ H _ _ _ _ O
u 3
l r O
L . 2
III.III .IIIIIII.
III AI
1,
fl ..
i.
_ _ _ _ H _ _
4 3 I 4 3 2 l

FRACTION NUMBER

 

,III‘ I i I III. III

 

99

FIGURE 8.--The Effect of Binding 35S-Labeled RNA Polymerase
to Poly(C) and Poly(A) on the Release of the Sigma Subunit
of RNA Polymerase.

Reaction mixtures (0.12 ml) which contained 100 mM Tris-HCl
(pH 8.0), 1 mM dithiothreitol, 10 ug/ml 358-labeled RNA
polymerase and either 2 mM MnC12, 250 ug/ml bovine serum
albumin, and 100 uM poly(C) (upper diagram), or 250 ug/ml
bovine serum albumin and 100 uM poly(C) (middle diagram),
or 2 mM MnClz and 100 uM poly(A) (lower diagram), were
incubated at 30° for 3 minutes and cooled. A sample (0.10
ml) from each reaction mixture was analyzed on sucrose
gradients containing 50 mM Tris-HCl (pH 8.0), 50 mM KCl,

1 mM dithiothreitol, and either 2 mM MnC12 and 100 ug/ml
bovine serum albumin (upper diagram), or 100 ug/ml bovine
serum albumin (middle diagram), or 2 mM MnClz (lower diagram)
as described in the legend to Figure 7.

Fraction number 11 (inset A), number 16 (inset B) and number
24 (inset C) from a duplicate sucrose gradient of a poly(A)-
containing reaction mixture were heated with SDS as described
in the legend to Figure 3 and then layered on 11 cm SDS—
polyacrylamide gels. Electrophoresis was at 25° at 3.6

ma/gel for 10 hr. After electrophoresis, the gels were cut
‘ into 4 mm transverse fractions and the 358 content of each
fraction was analyzed according to the procedure described
in Materials and Methods. For each inset, the ordinate is
expressed as 358—CPM x 10'2 and the abscissa as Gel Fraction
Number.

100

 

 

 

3

 

N
I
l

 

 

 

S-CPM X IO

35

 

 

 

 

 

 

 

 

 

 

l

0 IO 20 3O
FRACTION NUMBER

101

.nonEnz noHnomnm How on mmmHomno onu pnm NIOH x

SQUImmm mm commonmxo mH oumanno onu .uomnH onu non .mponnoS Una
mHMHnoHMS nH poanomop onncouonm onu on mannoooo powmeno mmB
nOHnomnm Hom nooo mo unonnoo mmm onn pnm mnoHnomnm omno>mnmnn EE v
ounH use mmz Hom onn .mHmononmonnooHo nonmd .nn mm.m noH Hom

\mE v no 0mm no mm3 mHmononmonuoon .Hom opHEMHSnooSHOQImom E0

HH no no ponoSMH nonn Uno m onanm on pnomoH onn nH poanomop

mo mom nuH3 ponmonn mma onnanE nOHnomon dzo HInm ponnnmnop

m we unoHponm omononm onm0HHdnp o Eonw v nonEnn nOHnoonm

.nonon HomZm ooanm m nH Emn ooo.om no nn m.m noH
pomannnnoo mmB mneIdOpgmHom manHonnoo unoHpmnm omononm onn
nonu noHnmooxo onu nnH3 m onanm on pnoooH onn nH poanomop
mm HUS SE om manHonnoo mnnoHpmnm omononm no poueHmno

mm3 onnanE nOHnomon nomo Eonm AHE OH.oO onEmm d .poHHHno
nonn one mounnHE m now com um pononnonH ono3 nEoanHp uannO
mmcnmssHoc azm cmHmncHImmm Hs\ma RH ccc .HneuaOcOnHoc z:

OOH .mHonS SE N no AEmnmoHp uHoHO ommnoEmHom dzm poHoanImmm
stcc mm ccc «zc HIcO,cmncecccc HE\O; OOH .Hon 25 ON .NHOcz
:5 H .NHocz :5 H ncruHc ccc HouHmnruoHcpHc as H ccc AO.O EEO
HonImHne SE om ponHmnnoo noHn3 AHE NH.oO monnanE noHnomom

.ommnoEwHom dzm mo ansnnm oEmHm
onn mo omooHom onn no mHeIdvpgmHom on one dzo HInm ponnumnoo

on omcnmanom dzm cmHmncHImmm cchch no ncmmnm oceII.m mmson

ll||llll|II

I0

102

N _

 

I

I I

 

 

 

 

 

 

 

 

 

l I I

 

 

 

VMN

E_0I x was-s“

FRACTION NUMBER

II I I1: 1' .I l

103

alfal- lllcl. Illl
Jul _.l| I‘ll .

.AOOOH .ncnmc snuoaonuoonm coHecHHHucHoo oHsoHH

Sn mmm now pommmmm nonn mo3 noHuomnm nomm .monnn omnHHnunoo onu
Ho Eonnon onn Eonm ponooHHoo ono3 HHE OH.OO mnOHnomnH .nonon
Homzm oocHnm one cH nan OO0.0m no no ~.m now as no coHuoocn
IHnnnoo nonwd .nHEnnHo Ennom onH>on HE\m: OOH one ~«HonS SE m
.HooHonEuoHnuHc :5 H .HOE :5 Om .A0.0 EEO HomImHne :5 om cH conon
Iona mos omononm one .omononm wom Ho HE O.O we now no unoHpmnm
omononm nmonHH HON on m HE m.v m manoSmH Sn ponmmonm unoHpmnm
omononm o no ponomoH mo3 onnanE nOHnomon nooo Eonw AHE OH.OV
oHnEmm d .poHHHno nonn pnm com um mopnnHE m now pononnonH ono3
neonoch EEOHnO nuoOSHon z: OmH no AaonooHo uHoHO AeoOnHon :5 mHH
nonnHo pno .ommnoESHom dzm poHonoHImmm HE\m: OH .nHEnnHm Ennom
ocH>on Ha\o: omm .NHocz as m .HouHonceochHo as H .AO.O moo
HomImHne SE OOH ponHmnnoo nOHn3 HHE NH.OO monnuxHE nOHuomom

.ommnoESHom
42E no eHccncm onOHm one no omcoHom one co nocOnHoo oco neoOnHoo
on omenoenHoo «2m ooHonoHImmm mchch no noonem oneII.OH EEOOHE

.ll iIl'l.l'

104

 

Om

ON

 

 

mums—Dz zo_._.o<mn.

O. on
_

 

 

ON

0.

N
e_0I x woo-s“

 

 

3
0
7
2

R N
"1
”6
”0
|Hh.3
“0
III! . .I3
Illa.
"2
ll 1
m3